#SampleID	BarcodeSequence	LinkerPrimerSequence	LAB_PERSON_CONTACT	SAMP_COLLECT_DEVICE	PROJECT_SPECIFIC_SITE	SOURCE_ENV	Env	SourceSink	KELLEY_OFFICE_SITE	ASSIGNED_FROM_GEO	RUN_DATE	TITLE	COMMON_SAMPLE_SITE	RUN_PREFIX	AGE	INVESTIGATION_TYPE	ENVIRONMENT	ANATOMICAL_SAMPLE_SITE	STATION	TOT_ORG_CARB	INCLUDES_TIMESERIES	BODY_SITE	SPECIFIC_LOCATION	ELEVATION	COLLECTION_DATE	GENDER	ALTITUDE	NUCL_ACID_AMP	PCR_COND	SEX	CARB_NITRO_RATIO	ANNUAL_SEASON_TEMP	BACTERIALCELLSPERM3	PRINCIPAL_INVESTIGATOR_CONTACT	HOST_SUBJECT_ID	ANONYMIZED_NAME	POOL_PROPORTION	SAMP_SIZE	FLOOR	URL	STUDY_DESCRIPTION	CMIN_RATE	VENTILATION_TYPE	LONGITUDE	MIENS_COMPLIANT	TOT_ORG_NITRO	STUDY_CENTER	NUCL_ACID_EXT	BODY_HABITAT	BODY_PRODUCT	STUDY_ABSTRACT	ENV_FEATURE	KEY_SEQ	STUDY_TITLE	BARCODE_READ_GROUP_TAG	REGION	EXPERIMENT_DESIGN_DESCRIPTION	PCR_PRIMERS	PMID	AGE_IN_YEARS	HOST_INDIVIDUAL	ORIGINAL_SAMPLE_SITE	LAB_PERSON	EXPERIMENT_CENTER	COUNTRY	TEXTURE	EXPERIMENT_ALIAS	DEPTH	TREATMENT	HOST_TAXID	SILT_CLAY	TEMP_UNITS	SUBMIT_TO_INSDC	NICU	PRINCIPAL_INVESTIGATOR	PROJECT_NAME	SEQUENCING_METH	ROOM	TEMP	SOIL_MOISTURE_DEFICIT	COMMON_NAME	LEVEL	ENV_BIOME	PLATFORM	RUN_ALIAS	AIRFLOWVELOCITYMPERSEC	ANNUAL_SEASON_PRECPT	PH	POOL_MEMBER_ACCESSION	STUDY_ALIAS	EXPERIMENT_TITLE	TAXON_ID	SAMPLE_CENTER	SOIL_TYPE	AGE_UNIT	AIRCHANGESPERHOUR	STUDY_ID	LIBRARY_CONSTRUCTION_PROTOCOL	STUDY_REF	LOCAL_TIME	PUBLIC	HOST_COMMON_NAME	POOL_MEMBER_NAME	ENV_MATTER	TARGET_GENE	BUILDING	VENTILATION_METHOD	TARGET_SUBFRAGMENT	RUN_CENTER	PRIMER_READ_GROUP_TAG	SURFACE	HUMIDITY	INSTRUMENT_NAME	HUMIDITY_UNITS	LATITUDE	TIME_ZONE	HOSPITAL	Description
M1.489861	AACCAAGG	CAAGAGTTTGATCCTGGCTCAG	skembel@uoregon.edu	NA	Indoor air	Indoor_air	NA	NA	NA	n	2012	Architectural design influences the diversity and structure of the built environment microbiome	NA	hospital_air_seqs	NA	mimarks-survey	Indoor	NA	NA	NA	0	NA	NA	9.94	2/27/12	NA	0	NA	NA	NA	NA	NA	1420000	jlgreen@uoregon.edu	NA	M1	NA	".1,g"	NA	NA	"Buildings are complex ecosystems that house trillions of microorganisms interacting with each other, with humans and with their environment. Understanding the ecological and evolutionary processes that determine the diversity and composition of the built environment microbiome???the community of microorganisms that live indoors???is important for understanding the relationship between building design, biodiversity and human health. In this study, we used high-throughput sequencing of the bacterial 16S rRNA gene to quantify relationships between building attributes and airborne bacterial communities at a health-care facility. We quantified airborne bacterial community structure and environmental conditions in patient rooms exposed to mechanical or window ventilation and in outdoor air. The phylogenetic diversity of airborne bacterial communities was lower indoors than outdoors, and mechanically ventilated rooms contained less diverse microbial communities than did window-ventilated rooms. Bacterial communities in indoor environments contained many taxa that are absent or rare outdoors, including taxa closely related to potential human pathogens. Building attributes, specifically the source of ventilation air, airflow rates, relative humidity and temperature, were correlated with the diversity and composition of indoor bacterial communities. The relative abundance of bacteria closely related to human pathogens was higher indoors than outdoors, and higher in rooms with lower airflow rates and lower relative humidity. The observed relationship between building design and airborne bacterial diversity suggests that we can manage indoor environments, altering through building design and operation the community of microbial species that potentially colonize the human microbiome during our time indoors."	NA	AC	-122.68	y	NA	U Oregon	NA	NA	NA	"Buildings are complex ecosystems that house trillions of microorganisms interacting with each other, with humans and with their environment. Understanding the ecological and evolutionary processes that determine the diversity and composition of the built environment microbiome the community of microorganisms that live indoors is important for understanding the relationship between building design, biodiversity and human health. In this study, we used high-throughput sequencing of the bacterial 16S rRNA gene to quantify relationships between building attributes and airborne bacterial communities at a health-care facility. We quantified airborne bacterial community structure and environmental conditions in patient rooms exposed to mechanical or window ventilation and in outdoor air. The phylogenetic diversity of airborne bacterial communities was lower indoors than outdoors, and mechanically ventilated rooms contained less diverse microbial communities than did window-ventilated rooms. Bacterial communities in indoor environments contained many taxa that are absent or rare outdoors, including taxa closely related to potential human pathogens. Building attributes, specifically the source of ventilation air, airflow rates, relative humidity and temperature, were correlated with the diversity and composition of indoor bacterial communities. The relative abundance of bacteria closely related to human pathogens was higher indoors than outdoors, and higher in rooms with lower airflow rates and lower relative humidity. The observed relationship between building design and airborne bacterial diversity suggests that we can manage indoor environments, altering through building design and operation the community of microbial species that potentially colonize the human microbiome during our time indoors."	ENVO:building	TCAG	Architectural design influences the diversity and structure of the built environment microbiome	NA	NA	high-throughput sequencing of the bacterial 16S rRNA gene to quantify relationships between building attributes and airborne bacterial communities at a health-care facility.	FWD:AGAGTTTGATCCTGGCTCAG; REV:TGCTGCCTCCCGTAGGAGT	PMID:22278670	NA	NA	NA	Steven Kembel/Gail Ackermann	U Oregon	GAZ:United States of America	NA	NA	0	Mechanical	NA	NA	C	n	NA	Jessica Green	Green_hospital_air	pyrosequencing	229	20.88706739	NA	air metagenome	NA	ENVO:city	FASTA	NA	9.016393443	NA	NA	NA	green_hospital_air_sloan	Architectural design influences the diversity and structure of the built environment microbiome	655179	U Oregon	NA	NA	5.358265096	1345	"FLX Titanium 27F, 338R"	NA	1130	y	NA	NA	ENVO:air	16S rRNA	NA	Indoor Mechanical	V2	Engencore	NA	NA	33.32786885	NA	percent	45.52	PST	NA	"mechanically ventilated hospital room, 11:30"
M2.489854	CCAAGGAA	CAAGAGTTTGATCCTGGCTCAG	skembel@uoregon.edu	NA	Indoor air	Indoor_air	NA	NA	NA	n	2012	Architectural design influences the diversity and structure of the built environment microbiome	NA	hospital_air_seqs	NA	mimarks-survey	Indoor	NA	NA	NA	0	NA	NA	9.94	2/27/12	NA	0	NA	NA	NA	NA	NA	937000	jlgreen@uoregon.edu	NA	M2	NA	".1,g"	NA	NA	"Buildings are complex ecosystems that house trillions of microorganisms interacting with each other, with humans and with their environment. Understanding the ecological and evolutionary processes that determine the diversity and composition of the built environment microbiome???the community of microorganisms that live indoors???is important for understanding the relationship between building design, biodiversity and human health. In this study, we used high-throughput sequencing of the bacterial 16S rRNA gene to quantify relationships between building attributes and airborne bacterial communities at a health-care facility. We quantified airborne bacterial community structure and environmental conditions in patient rooms exposed to mechanical or window ventilation and in outdoor air. The phylogenetic diversity of airborne bacterial communities was lower indoors than outdoors, and mechanically ventilated rooms contained less diverse microbial communities than did window-ventilated rooms. Bacterial communities in indoor environments contained many taxa that are absent or rare outdoors, including taxa closely related to potential human pathogens. Building attributes, specifically the source of ventilation air, airflow rates, relative humidity and temperature, were correlated with the diversity and composition of indoor bacterial communities. The relative abundance of bacteria closely related to human pathogens was higher indoors than outdoors, and higher in rooms with lower airflow rates and lower relative humidity. The observed relationship between building design and airborne bacterial diversity suggests that we can manage indoor environments, altering through building design and operation the community of microbial species that potentially colonize the human microbiome during our time indoors."	NA	AC	-122.68	y	NA	U Oregon	NA	NA	NA	"Buildings are complex ecosystems that house trillions of microorganisms interacting with each other, with humans and with their environment. Understanding the ecological and evolutionary processes that determine the diversity and composition of the built environment microbiome the community of microorganisms that live indoors is important for understanding the relationship between building design, biodiversity and human health. In this study, we used high-throughput sequencing of the bacterial 16S rRNA gene to quantify relationships between building attributes and airborne bacterial communities at a health-care facility. We quantified airborne bacterial community structure and environmental conditions in patient rooms exposed to mechanical or window ventilation and in outdoor air. The phylogenetic diversity of airborne bacterial communities was lower indoors than outdoors, and mechanically ventilated rooms contained less diverse microbial communities than did window-ventilated rooms. Bacterial communities in indoor environments contained many taxa that are absent or rare outdoors, including taxa closely related to potential human pathogens. Building attributes, specifically the source of ventilation air, airflow rates, relative humidity and temperature, were correlated with the diversity and composition of indoor bacterial communities. The relative abundance of bacteria closely related to human pathogens was higher indoors than outdoors, and higher in rooms with lower airflow rates and lower relative humidity. The observed relationship between building design and airborne bacterial diversity suggests that we can manage indoor environments, altering through building design and operation the community of microbial species that potentially colonize the human microbiome during our time indoors."	ENVO:building	TCAG	Architectural design influences the diversity and structure of the built environment microbiome	NA	NA	high-throughput sequencing of the bacterial 16S rRNA gene to quantify relationships between building attributes and airborne bacterial communities at a health-care facility.	FWD:AGAGTTTGATCCTGGCTCAG; REV:TGCTGCCTCCCGTAGGAGT	PMID:22278670	NA	NA	NA	Steven Kembel/Gail Ackermann	U Oregon	GAZ:United States of America	NA	NA	0	Mechanical	NA	NA	C	n	NA	Jessica Green	Green_hospital_air	pyrosequencing	229	21.29690346	NA	air metagenome	NA	ENVO:city	FASTA	NA	8.655737705	NA	NA	NA	green_hospital_air_sloan	Architectural design influences the diversity and structure of the built environment microbiome	655179	U Oregon	NA	NA	3.529165064	1345	"FLX Titanium 27F, 338R"	NA	1300	y	NA	NA	ENVO:air	16S rRNA	NA	Indoor Mechanical	V2	Engencore	NA	NA	32.86229508	NA	percent	45.52	PST	NA	"mechanically ventilated hospital room, 13:00"
M3.489857	GAAGACCA	CAAGAGTTTGATCCTGGCTCAG	skembel@uoregon.edu	NA	Indoor air	Indoor_air	NA	NA	NA	n	2012	Architectural design influences the diversity and structure of the built environment microbiome	NA	hospital_air_seqs	NA	mimarks-survey	Indoor	NA	NA	NA	0	NA	NA	9.94	2/27/12	NA	0	NA	NA	NA	NA	NA	607000	jlgreen@uoregon.edu	NA	M3	NA	".1,g"	NA	NA	"Buildings are complex ecosystems that house trillions of microorganisms interacting with each other, with humans and with their environment. Understanding the ecological and evolutionary processes that determine the diversity and composition of the built environment microbiome???the community of microorganisms that live indoors???is important for understanding the relationship between building design, biodiversity and human health. In this study, we used high-throughput sequencing of the bacterial 16S rRNA gene to quantify relationships between building attributes and airborne bacterial communities at a health-care facility. We quantified airborne bacterial community structure and environmental conditions in patient rooms exposed to mechanical or window ventilation and in outdoor air. The phylogenetic diversity of airborne bacterial communities was lower indoors than outdoors, and mechanically ventilated rooms contained less diverse microbial communities than did window-ventilated rooms. Bacterial communities in indoor environments contained many taxa that are absent or rare outdoors, including taxa closely related to potential human pathogens. Building attributes, specifically the source of ventilation air, airflow rates, relative humidity and temperature, were correlated with the diversity and composition of indoor bacterial communities. The relative abundance of bacteria closely related to human pathogens was higher indoors than outdoors, and higher in rooms with lower airflow rates and lower relative humidity. The observed relationship between building design and airborne bacterial diversity suggests that we can manage indoor environments, altering through building design and operation the community of microbial species that potentially colonize the human microbiome during our time indoors."	NA	AC	-122.68	y	NA	U Oregon	NA	NA	NA	"Buildings are complex ecosystems that house trillions of microorganisms interacting with each other, with humans and with their environment. Understanding the ecological and evolutionary processes that determine the diversity and composition of the built environment microbiome the community of microorganisms that live indoors is important for understanding the relationship between building design, biodiversity and human health. In this study, we used high-throughput sequencing of the bacterial 16S rRNA gene to quantify relationships between building attributes and airborne bacterial communities at a health-care facility. We quantified airborne bacterial community structure and environmental conditions in patient rooms exposed to mechanical or window ventilation and in outdoor air. The phylogenetic diversity of airborne bacterial communities was lower indoors than outdoors, and mechanically ventilated rooms contained less diverse microbial communities than did window-ventilated rooms. Bacterial communities in indoor environments contained many taxa that are absent or rare outdoors, including taxa closely related to potential human pathogens. Building attributes, specifically the source of ventilation air, airflow rates, relative humidity and temperature, were correlated with the diversity and composition of indoor bacterial communities. The relative abundance of bacteria closely related to human pathogens was higher indoors than outdoors, and higher in rooms with lower airflow rates and lower relative humidity. The observed relationship between building design and airborne bacterial diversity suggests that we can manage indoor environments, altering through building design and operation the community of microbial species that potentially colonize the human microbiome during our time indoors."	ENVO:building	TCAG	Architectural design influences the diversity and structure of the built environment microbiome	NA	NA	high-throughput sequencing of the bacterial 16S rRNA gene to quantify relationships between building attributes and airborne bacterial communities at a health-care facility.	FWD:AGAGTTTGATCCTGGCTCAG; REV:TGCTGCCTCCCGTAGGAGT	PMID:22278670	NA	NA	NA	Steven Kembel/Gail Ackermann	U Oregon	GAZ:United States of America	NA	NA	0	Mechanical	NA	NA	C	n	NA	Jessica Green	Green_hospital_air	pyrosequencing	231	22.25434028	NA	air metagenome	NA	ENVO:city	FASTA	NA	16.140625	NA	NA	NA	green_hospital_air_sloan	Architectural design influences the diversity and structure of the built environment microbiome	655179	U Oregon	NA	NA	8.406254985	1345	"FLX Titanium 27F, 338R"	NA	1600	y	NA	NA	ENVO:air	16S rRNA	NA	Indoor Mechanical	V2	Engencore	NA	NA	31.6703125	NA	percent	45.52	PST	NA	"mechanically ventilated hospital room, 16:00"
M4.489862	GCTTGCTT	CAAGAGTTTGATCCTGGCTCAG	skembel@uoregon.edu	NA	Indoor air	Indoor_air	NA	NA	NA	n	2012	Architectural design influences the diversity and structure of the built environment microbiome	NA	hospital_air_seqs	NA	mimarks-survey	Indoor	NA	NA	NA	0	NA	NA	9.94	2/27/12	NA	0	NA	NA	NA	NA	NA	1200000	jlgreen@uoregon.edu	NA	M4	NA	".1,g"	NA	NA	"Buildings are complex ecosystems that house trillions of microorganisms interacting with each other, with humans and with their environment. Understanding the ecological and evolutionary processes that determine the diversity and composition of the built environment microbiome???the community of microorganisms that live indoors???is important for understanding the relationship between building design, biodiversity and human health. In this study, we used high-throughput sequencing of the bacterial 16S rRNA gene to quantify relationships between building attributes and airborne bacterial communities at a health-care facility. We quantified airborne bacterial community structure and environmental conditions in patient rooms exposed to mechanical or window ventilation and in outdoor air. The phylogenetic diversity of airborne bacterial communities was lower indoors than outdoors, and mechanically ventilated rooms contained less diverse microbial communities than did window-ventilated rooms. Bacterial communities in indoor environments contained many taxa that are absent or rare outdoors, including taxa closely related to potential human pathogens. Building attributes, specifically the source of ventilation air, airflow rates, relative humidity and temperature, were correlated with the diversity and composition of indoor bacterial communities. The relative abundance of bacteria closely related to human pathogens was higher indoors than outdoors, and higher in rooms with lower airflow rates and lower relative humidity. The observed relationship between building design and airborne bacterial diversity suggests that we can manage indoor environments, altering through building design and operation the community of microbial species that potentially colonize the human microbiome during our time indoors."	NA	AC	-122.68	y	NA	U Oregon	NA	NA	NA	"Buildings are complex ecosystems that house trillions of microorganisms interacting with each other, with humans and with their environment. Understanding the ecological and evolutionary processes that determine the diversity and composition of the built environment microbiome the community of microorganisms that live indoors is important for understanding the relationship between building design, biodiversity and human health. In this study, we used high-throughput sequencing of the bacterial 16S rRNA gene to quantify relationships between building attributes and airborne bacterial communities at a health-care facility. We quantified airborne bacterial community structure and environmental conditions in patient rooms exposed to mechanical or window ventilation and in outdoor air. The phylogenetic diversity of airborne bacterial communities was lower indoors than outdoors, and mechanically ventilated rooms contained less diverse microbial communities than did window-ventilated rooms. Bacterial communities in indoor environments contained many taxa that are absent or rare outdoors, including taxa closely related to potential human pathogens. Building attributes, specifically the source of ventilation air, airflow rates, relative humidity and temperature, were correlated with the diversity and composition of indoor bacterial communities. The relative abundance of bacteria closely related to human pathogens was higher indoors than outdoors, and higher in rooms with lower airflow rates and lower relative humidity. The observed relationship between building design and airborne bacterial diversity suggests that we can manage indoor environments, altering through building design and operation the community of microbial species that potentially colonize the human microbiome during our time indoors."	ENVO:building	TCAG	Architectural design influences the diversity and structure of the built environment microbiome	NA	NA	high-throughput sequencing of the bacterial 16S rRNA gene to quantify relationships between building attributes and airborne bacterial communities at a health-care facility.	FWD:AGAGTTTGATCCTGGCTCAG; REV:TGCTGCCTCCCGTAGGAGT	PMID:22278670	NA	NA	NA	Steven Kembel/Gail Ackermann	U Oregon	GAZ:United States of America	NA	NA	0	Mechanical	NA	NA	C	n	NA	Jessica Green	Green_hospital_air	pyrosequencing	231	21.91093474	NA	air metagenome	NA	ENVO:city	FASTA	NA	20.77777778	NA	NA	NA	green_hospital_air_sloan	Architectural design influences the diversity and structure of the built environment microbiome	655179	U Oregon	NA	NA	8.512498341	1345	"FLX Titanium 27F, 338R"	NA	1700	y	NA	NA	ENVO:air	16S rRNA	NA	Indoor Mechanical	V2	Engencore	NA	NA	32.65555556	NA	percent	45.52	PST	NA	"mechanically ventilated hospital room, 17:00"
M5.489864	TCCTCTTC	CAAGAGTTTGATCCTGGCTCAG	skembel@uoregon.edu	NA	Indoor air	Indoor_air	NA	NA	NA	n	2012	Architectural design influences the diversity and structure of the built environment microbiome	NA	hospital_air_seqs	NA	mimarks-survey	Indoor	NA	NA	NA	0	NA	NA	9.94	2/28/12 12:20	NA	0	NA	NA	NA	NA	NA	2580000	jlgreen@uoregon.edu	NA	M5	NA	".1,g"	NA	NA	"Buildings are complex ecosystems that house trillions of microorganisms interacting with each other, with humans and with their environment. Understanding the ecological and evolutionary processes that determine the diversity and composition of the built environment microbiome???the community of microorganisms that live indoors???is important for understanding the relationship between building design, biodiversity and human health. In this study, we used high-throughput sequencing of the bacterial 16S rRNA gene to quantify relationships between building attributes and airborne bacterial communities at a health-care facility. We quantified airborne bacterial community structure and environmental conditions in patient rooms exposed to mechanical or window ventilation and in outdoor air. The phylogenetic diversity of airborne bacterial communities was lower indoors than outdoors, and mechanically ventilated rooms contained less diverse microbial communities than did window-ventilated rooms. Bacterial communities in indoor environments contained many taxa that are absent or rare outdoors, including taxa closely related to potential human pathogens. Building attributes, specifically the source of ventilation air, airflow rates, relative humidity and temperature, were correlated with the diversity and composition of indoor bacterial communities. The relative abundance of bacteria closely related to human pathogens was higher indoors than outdoors, and higher in rooms with lower airflow rates and lower relative humidity. The observed relationship between building design and airborne bacterial diversity suggests that we can manage indoor environments, altering through building design and operation the community of microbial species that potentially colonize the human microbiome during our time indoors."	NA	AC	-122.68	y	NA	U Oregon	NA	NA	NA	"Buildings are complex ecosystems that house trillions of microorganisms interacting with each other, with humans and with their environment. Understanding the ecological and evolutionary processes that determine the diversity and composition of the built environment microbiome the community of microorganisms that live indoors is important for understanding the relationship between building design, biodiversity and human health. In this study, we used high-throughput sequencing of the bacterial 16S rRNA gene to quantify relationships between building attributes and airborne bacterial communities at a health-care facility. We quantified airborne bacterial community structure and environmental conditions in patient rooms exposed to mechanical or window ventilation and in outdoor air. The phylogenetic diversity of airborne bacterial communities was lower indoors than outdoors, and mechanically ventilated rooms contained less diverse microbial communities than did window-ventilated rooms. Bacterial communities in indoor environments contained many taxa that are absent or rare outdoors, including taxa closely related to potential human pathogens. Building attributes, specifically the source of ventilation air, airflow rates, relative humidity and temperature, were correlated with the diversity and composition of indoor bacterial communities. The relative abundance of bacteria closely related to human pathogens was higher indoors than outdoors, and higher in rooms with lower airflow rates and lower relative humidity. The observed relationship between building design and airborne bacterial diversity suggests that we can manage indoor environments, altering through building design and operation the community of microbial species that potentially colonize the human microbiome during our time indoors."	ENVO:building	TCAG	Architectural design influences the diversity and structure of the built environment microbiome	NA	NA	high-throughput sequencing of the bacterial 16S rRNA gene to quantify relationships between building attributes and airborne bacterial communities at a health-care facility.	FWD:AGAGTTTGATCCTGGCTCAG; REV:TGCTGCCTCCCGTAGGAGT	PMID:22278670	NA	NA	NA	Steven Kembel/Gail Ackermann	U Oregon	GAZ:United States of America	NA	NA	0	Mechanical	NA	NA	C	n	NA	Jessica Green	Green_hospital_air	pyrosequencing	235	22.98148148	NA	air metagenome	NA	ENVO:city	FASTA	NA	4.484848485	NA	NA	NA	green_hospital_air_sloan	Architectural design influences the diversity and structure of the built environment microbiome	655179	U Oregon	NA	NA	5.102859082	1345	"FLX Titanium 27F, 338R"	NA	1220	y	NA	NA	ENVO:air	16S rRNA	NA	Indoor Mechanical	V2	Engencore	NA	NA	32.82575758	NA	percent	45.52	PST	NA	"mechanically ventilated hospital room, 12:20, 2/28/2012"
W1.489853	AACCAACC	CAAGAGTTTGATCCTGGCTCAG	skembel@uoregon.edu	NA	Indoor air	Indoor_air	NA	NA	NA	n	2012	Architectural design influences the diversity and structure of the built environment microbiome	NA	hospital_air_seqs	NA	mimarks-survey	Indoor	NA	NA	NA	0	NA	NA	9.94	2/27/12	NA	0	NA	NA	NA	NA	NA	723000	jlgreen@uoregon.edu	NA	W1	NA	".1,g"	NA	NA	"Buildings are complex ecosystems that house trillions of microorganisms interacting with each other, with humans and with their environment. Understanding the ecological and evolutionary processes that determine the diversity and composition of the built environment microbiome???the community of microorganisms that live indoors???is important for understanding the relationship between building design, biodiversity and human health. In this study, we used high-throughput sequencing of the bacterial 16S rRNA gene to quantify relationships between building attributes and airborne bacterial communities at a health-care facility. We quantified airborne bacterial community structure and environmental conditions in patient rooms exposed to mechanical or window ventilation and in outdoor air. The phylogenetic diversity of airborne bacterial communities was lower indoors than outdoors, and mechanically ventilated rooms contained less diverse microbial communities than did window-ventilated rooms. Bacterial communities in indoor environments contained many taxa that are absent or rare outdoors, including taxa closely related to potential human pathogens. Building attributes, specifically the source of ventilation air, airflow rates, relative humidity and temperature, were correlated with the diversity and composition of indoor bacterial communities. The relative abundance of bacteria closely related to human pathogens was higher indoors than outdoors, and higher in rooms with lower airflow rates and lower relative humidity. The observed relationship between building design and airborne bacterial diversity suggests that we can manage indoor environments, altering through building design and operation the community of microbial species that potentially colonize the human microbiome during our time indoors."	NA	window	-122.68	y	NA	U Oregon	NA	NA	NA	"Buildings are complex ecosystems that house trillions of microorganisms interacting with each other, with humans and with their environment. Understanding the ecological and evolutionary processes that determine the diversity and composition of the built environment microbiome the community of microorganisms that live indoors is important for understanding the relationship between building design, biodiversity and human health. In this study, we used high-throughput sequencing of the bacterial 16S rRNA gene to quantify relationships between building attributes and airborne bacterial communities at a health-care facility. We quantified airborne bacterial community structure and environmental conditions in patient rooms exposed to mechanical or window ventilation and in outdoor air. The phylogenetic diversity of airborne bacterial communities was lower indoors than outdoors, and mechanically ventilated rooms contained less diverse microbial communities than did window-ventilated rooms. Bacterial communities in indoor environments contained many taxa that are absent or rare outdoors, including taxa closely related to potential human pathogens. Building attributes, specifically the source of ventilation air, airflow rates, relative humidity and temperature, were correlated with the diversity and composition of indoor bacterial communities. The relative abundance of bacteria closely related to human pathogens was higher indoors than outdoors, and higher in rooms with lower airflow rates and lower relative humidity. The observed relationship between building design and airborne bacterial diversity suggests that we can manage indoor environments, altering through building design and operation the community of microbial species that potentially colonize the human microbiome during our time indoors."	ENVO:building	TCAG	Architectural design influences the diversity and structure of the built environment microbiome	NA	NA	high-throughput sequencing of the bacterial 16S rRNA gene to quantify relationships between building attributes and airborne bacterial communities at a health-care facility.	FWD:AGAGTTTGATCCTGGCTCAG; REV:TGCTGCCTCCCGTAGGAGT	PMID:22278670	NA	NA	NA	Steven Kembel/Gail Ackermann	U Oregon	GAZ:United States of America	NA	NA	0	Window	NA	NA	C	n	NA	Jessica Green	Green_hospital_air	pyrosequencing	235	20.26451078	NA	air metagenome	NA	ENVO:city	FASTA	NA	21.05970149	NA	NA	NA	green_hospital_air_sloan	Architectural design influences the diversity and structure of the built environment microbiome	655179	U Oregon	NA	NA	11.71536064	1345	"FLX Titanium 27F, 338R"	NA	1130	y	NA	NA	ENVO:air	16S rRNA	NA	Indoor Window	V2	Engencore	NA	NA	34.76567164	NA	percent	45.52	PST	NA	"window ventilated hospital room, 11:30"
W3.489863	GAACACCT	CAAGAGTTTGATCCTGGCTCAG	skembel@uoregon.edu	NA	Indoor air	Indoor_air	NA	NA	NA	n	2012	Architectural design influences the diversity and structure of the built environment microbiome	NA	hospital_air_seqs	NA	mimarks-survey	Indoor	NA	NA	NA	0	NA	NA	9.94	2/27/12	NA	0	NA	NA	NA	NA	NA	867000	jlgreen@uoregon.edu	NA	W3	NA	".1,g"	NA	NA	"Buildings are complex ecosystems that house trillions of microorganisms interacting with each other, with humans and with their environment. Understanding the ecological and evolutionary processes that determine the diversity and composition of the built environment microbiome???the community of microorganisms that live indoors???is important for understanding the relationship between building design, biodiversity and human health. In this study, we used high-throughput sequencing of the bacterial 16S rRNA gene to quantify relationships between building attributes and airborne bacterial communities at a health-care facility. We quantified airborne bacterial community structure and environmental conditions in patient rooms exposed to mechanical or window ventilation and in outdoor air. The phylogenetic diversity of airborne bacterial communities was lower indoors than outdoors, and mechanically ventilated rooms contained less diverse microbial communities than did window-ventilated rooms. Bacterial communities in indoor environments contained many taxa that are absent or rare outdoors, including taxa closely related to potential human pathogens. Building attributes, specifically the source of ventilation air, airflow rates, relative humidity and temperature, were correlated with the diversity and composition of indoor bacterial communities. The relative abundance of bacteria closely related to human pathogens was higher indoors than outdoors, and higher in rooms with lower airflow rates and lower relative humidity. The observed relationship between building design and airborne bacterial diversity suggests that we can manage indoor environments, altering through building design and operation the community of microbial species that potentially colonize the human microbiome during our time indoors."	NA	window	-122.68	y	NA	U Oregon	NA	NA	NA	"Buildings are complex ecosystems that house trillions of microorganisms interacting with each other, with humans and with their environment. Understanding the ecological and evolutionary processes that determine the diversity and composition of the built environment microbiome the community of microorganisms that live indoors is important for understanding the relationship between building design, biodiversity and human health. In this study, we used high-throughput sequencing of the bacterial 16S rRNA gene to quantify relationships between building attributes and airborne bacterial communities at a health-care facility. We quantified airborne bacterial community structure and environmental conditions in patient rooms exposed to mechanical or window ventilation and in outdoor air. The phylogenetic diversity of airborne bacterial communities was lower indoors than outdoors, and mechanically ventilated rooms contained less diverse microbial communities than did window-ventilated rooms. Bacterial communities in indoor environments contained many taxa that are absent or rare outdoors, including taxa closely related to potential human pathogens. Building attributes, specifically the source of ventilation air, airflow rates, relative humidity and temperature, were correlated with the diversity and composition of indoor bacterial communities. The relative abundance of bacteria closely related to human pathogens was higher indoors than outdoors, and higher in rooms with lower airflow rates and lower relative humidity. The observed relationship between building design and airborne bacterial diversity suggests that we can manage indoor environments, altering through building design and operation the community of microbial species that potentially colonize the human microbiome during our time indoors."	ENVO:building	TCAG	Architectural design influences the diversity and structure of the built environment microbiome	NA	NA	high-throughput sequencing of the bacterial 16S rRNA gene to quantify relationships between building attributes and airborne bacterial communities at a health-care facility.	FWD:AGAGTTTGATCCTGGCTCAG; REV:TGCTGCCTCCCGTAGGAGT	PMID:22278670	NA	NA	NA	Steven Kembel/Gail Ackermann	U Oregon	GAZ:United States of America	NA	NA	0	Window	NA	NA	C	n	NA	Jessica Green	Green_hospital_air	pyrosequencing	229	23.18189964	NA	air metagenome	NA	ENVO:city	FASTA	NA	8.290322581	NA	NA	NA	green_hospital_air_sloan	Architectural design influences the diversity and structure of the built environment microbiome	655179	U Oregon	NA	NA	13.68490584	1345	"FLX Titanium 27F, 338R"	NA	1600	y	NA	NA	ENVO:air	16S rRNA	NA	Indoor Window	V2	Engencore	NA	NA	30.2516129	NA	percent	45.52	PST	NA	"window ventilated hospital room, 16:00"
W4.489865	GGCCTAAT	CAAGAGTTTGATCCTGGCTCAG	skembel@uoregon.edu	NA	Indoor air	Indoor_air	NA	NA	NA	n	2012	Architectural design influences the diversity and structure of the built environment microbiome	NA	hospital_air_seqs	NA	mimarks-survey	Indoor	NA	NA	NA	0	NA	NA	9.94	2/27/12	NA	0	NA	NA	NA	NA	NA	625000	jlgreen@uoregon.edu	NA	W4	NA	".1,g"	NA	NA	"Buildings are complex ecosystems that house trillions of microorganisms interacting with each other, with humans and with their environment. Understanding the ecological and evolutionary processes that determine the diversity and composition of the built environment microbiome???the community of microorganisms that live indoors???is important for understanding the relationship between building design, biodiversity and human health. In this study, we used high-throughput sequencing of the bacterial 16S rRNA gene to quantify relationships between building attributes and airborne bacterial communities at a health-care facility. We quantified airborne bacterial community structure and environmental conditions in patient rooms exposed to mechanical or window ventilation and in outdoor air. The phylogenetic diversity of airborne bacterial communities was lower indoors than outdoors, and mechanically ventilated rooms contained less diverse microbial communities than did window-ventilated rooms. Bacterial communities in indoor environments contained many taxa that are absent or rare outdoors, including taxa closely related to potential human pathogens. Building attributes, specifically the source of ventilation air, airflow rates, relative humidity and temperature, were correlated with the diversity and composition of indoor bacterial communities. The relative abundance of bacteria closely related to human pathogens was higher indoors than outdoors, and higher in rooms with lower airflow rates and lower relative humidity. The observed relationship between building design and airborne bacterial diversity suggests that we can manage indoor environments, altering through building design and operation the community of microbial species that potentially colonize the human microbiome during our time indoors."	NA	window	-122.68	y	NA	U Oregon	NA	NA	NA	"Buildings are complex ecosystems that house trillions of microorganisms interacting with each other, with humans and with their environment. Understanding the ecological and evolutionary processes that determine the diversity and composition of the built environment microbiome the community of microorganisms that live indoors is important for understanding the relationship between building design, biodiversity and human health. In this study, we used high-throughput sequencing of the bacterial 16S rRNA gene to quantify relationships between building attributes and airborne bacterial communities at a health-care facility. We quantified airborne bacterial community structure and environmental conditions in patient rooms exposed to mechanical or window ventilation and in outdoor air. The phylogenetic diversity of airborne bacterial communities was lower indoors than outdoors, and mechanically ventilated rooms contained less diverse microbial communities than did window-ventilated rooms. Bacterial communities in indoor environments contained many taxa that are absent or rare outdoors, including taxa closely related to potential human pathogens. Building attributes, specifically the source of ventilation air, airflow rates, relative humidity and temperature, were correlated with the diversity and composition of indoor bacterial communities. The relative abundance of bacteria closely related to human pathogens was higher indoors than outdoors, and higher in rooms with lower airflow rates and lower relative humidity. The observed relationship between building design and airborne bacterial diversity suggests that we can manage indoor environments, altering through building design and operation the community of microbial species that potentially colonize the human microbiome during our time indoors."	ENVO:building	TCAG	Architectural design influences the diversity and structure of the built environment microbiome	NA	NA	high-throughput sequencing of the bacterial 16S rRNA gene to quantify relationships between building attributes and airborne bacterial communities at a health-care facility.	FWD:AGAGTTTGATCCTGGCTCAG; REV:TGCTGCCTCCCGTAGGAGT	PMID:22278670	NA	NA	NA	Steven Kembel/Gail Ackermann	U Oregon	GAZ:United States of America	NA	NA	0	Window	NA	NA	C	n	NA	Jessica Green	Green_hospital_air	pyrosequencing	229	20.75308642	NA	air metagenome	NA	ENVO:city	FASTA	NA	8.80952381	NA	NA	NA	green_hospital_air_sloan	Architectural design influences the diversity and structure of the built environment microbiome	655179	U Oregon	NA	NA	12.05542727	1345	"FLX Titanium 27F, 338R"	NA	1700	y	NA	NA	ENVO:air	16S rRNA	NA	Indoor Window	V2	Engencore	NA	NA	34.96507937	NA	percent	45.52	PST	NA	"window ventilated hospital room, 17:00"
W5.489859	GTTGCTTC	CAAGAGTTTGATCCTGGCTCAG	skembel@uoregon.edu	NA	Indoor air	Indoor_air	NA	NA	NA	n	2012	Architectural design influences the diversity and structure of the built environment microbiome	NA	hospital_air_seqs	NA	mimarks-survey	Indoor	NA	NA	NA	0	NA	NA	9.94	2/28/12 12:20	NA	0	NA	NA	NA	NA	NA	969000	jlgreen@uoregon.edu	NA	W5	NA	".1,g"	NA	NA	"Buildings are complex ecosystems that house trillions of microorganisms interacting with each other, with humans and with their environment. Understanding the ecological and evolutionary processes that determine the diversity and composition of the built environment microbiome???the community of microorganisms that live indoors???is important for understanding the relationship between building design, biodiversity and human health. In this study, we used high-throughput sequencing of the bacterial 16S rRNA gene to quantify relationships between building attributes and airborne bacterial communities at a health-care facility. We quantified airborne bacterial community structure and environmental conditions in patient rooms exposed to mechanical or window ventilation and in outdoor air. The phylogenetic diversity of airborne bacterial communities was lower indoors than outdoors, and mechanically ventilated rooms contained less diverse microbial communities than did window-ventilated rooms. Bacterial communities in indoor environments contained many taxa that are absent or rare outdoors, including taxa closely related to potential human pathogens. Building attributes, specifically the source of ventilation air, airflow rates, relative humidity and temperature, were correlated with the diversity and composition of indoor bacterial communities. The relative abundance of bacteria closely related to human pathogens was higher indoors than outdoors, and higher in rooms with lower airflow rates and lower relative humidity. The observed relationship between building design and airborne bacterial diversity suggests that we can manage indoor environments, altering through building design and operation the community of microbial species that potentially colonize the human microbiome during our time indoors."	NA	window	-122.68	y	NA	U Oregon	NA	NA	NA	"Buildings are complex ecosystems that house trillions of microorganisms interacting with each other, with humans and with their environment. Understanding the ecological and evolutionary processes that determine the diversity and composition of the built environment microbiome the community of microorganisms that live indoors is important for understanding the relationship between building design, biodiversity and human health. In this study, we used high-throughput sequencing of the bacterial 16S rRNA gene to quantify relationships between building attributes and airborne bacterial communities at a health-care facility. We quantified airborne bacterial community structure and environmental conditions in patient rooms exposed to mechanical or window ventilation and in outdoor air. The phylogenetic diversity of airborne bacterial communities was lower indoors than outdoors, and mechanically ventilated rooms contained less diverse microbial communities than did window-ventilated rooms. Bacterial communities in indoor environments contained many taxa that are absent or rare outdoors, including taxa closely related to potential human pathogens. Building attributes, specifically the source of ventilation air, airflow rates, relative humidity and temperature, were correlated with the diversity and composition of indoor bacterial communities. The relative abundance of bacteria closely related to human pathogens was higher indoors than outdoors, and higher in rooms with lower airflow rates and lower relative humidity. The observed relationship between building design and airborne bacterial diversity suggests that we can manage indoor environments, altering through building design and operation the community of microbial species that potentially colonize the human microbiome during our time indoors."	ENVO:building	TCAG	Architectural design influences the diversity and structure of the built environment microbiome	NA	NA	high-throughput sequencing of the bacterial 16S rRNA gene to quantify relationships between building attributes and airborne bacterial communities at a health-care facility.	FWD:AGAGTTTGATCCTGGCTCAG; REV:TGCTGCCTCCCGTAGGAGT	PMID:22278670	NA	NA	NA	Steven Kembel/Gail Ackermann	U Oregon	GAZ:United States of America	NA	NA	0	Window	NA	NA	C	n	NA	Jessica Green	Green_hospital_air	pyrosequencing	231	21.61025641	NA	air metagenome	NA	ENVO:city	FASTA	NA	12.16923077	NA	NA	NA	green_hospital_air_sloan	Architectural design influences the diversity and structure of the built environment microbiome	655179	U Oregon	NA	NA	8.275441425	1345	"FLX Titanium 27F, 338R"	NA	1220	y	NA	NA	ENVO:air	16S rRNA	NA	Indoor Window	V2	Engencore	NA	NA	37.35230769	NA	percent	45.52	PST	NA	"window ventilated hospital room, 12:20, 2/28/2012"
O1.489860	CAACACCA	CAAGAGTTTGATCCTGGCTCAG	skembel@uoregon.edu	NA	Outdoor air	Outdoor_air	Outdoor Air	source	NA	n	2012	Architectural design influences the diversity and structure of the built environment microbiome	NA	hospital_air_seqs	NA	mimarks-survey	Outdoor	NA	NA	NA	0	NA	NA	9.94	2/27/12	NA	0	NA	NA	NA	NA	NA	572000	jlgreen@uoregon.edu	NA	O1	NA	".1,g"	NA	NA	"Buildings are complex ecosystems that house trillions of microorganisms interacting with each other, with humans and with their environment. Understanding the ecological and evolutionary processes that determine the diversity and composition of the built environment microbiome???the community of microorganisms that live indoors???is important for understanding the relationship between building design, biodiversity and human health. In this study, we used high-throughput sequencing of the bacterial 16S rRNA gene to quantify relationships between building attributes and airborne bacterial communities at a health-care facility. We quantified airborne bacterial community structure and environmental conditions in patient rooms exposed to mechanical or window ventilation and in outdoor air. The phylogenetic diversity of airborne bacterial communities was lower indoors than outdoors, and mechanically ventilated rooms contained less diverse microbial communities than did window-ventilated rooms. Bacterial communities in indoor environments contained many taxa that are absent or rare outdoors, including taxa closely related to potential human pathogens. Building attributes, specifically the source of ventilation air, airflow rates, relative humidity and temperature, were correlated with the diversity and composition of indoor bacterial communities. The relative abundance of bacteria closely related to human pathogens was higher indoors than outdoors, and higher in rooms with lower airflow rates and lower relative humidity. The observed relationship between building design and airborne bacterial diversity suggests that we can manage indoor environments, altering through building design and operation the community of microbial species that potentially colonize the human microbiome during our time indoors."	NA	"none, outside air"	-122.68	y	NA	U Oregon	NA	NA	NA	"Buildings are complex ecosystems that house trillions of microorganisms interacting with each other, with humans and with their environment. Understanding the ecological and evolutionary processes that determine the diversity and composition of the built environment microbiome the community of microorganisms that live indoors is important for understanding the relationship between building design, biodiversity and human health. In this study, we used high-throughput sequencing of the bacterial 16S rRNA gene to quantify relationships between building attributes and airborne bacterial communities at a health-care facility. We quantified airborne bacterial community structure and environmental conditions in patient rooms exposed to mechanical or window ventilation and in outdoor air. The phylogenetic diversity of airborne bacterial communities was lower indoors than outdoors, and mechanically ventilated rooms contained less diverse microbial communities than did window-ventilated rooms. Bacterial communities in indoor environments contained many taxa that are absent or rare outdoors, including taxa closely related to potential human pathogens. Building attributes, specifically the source of ventilation air, airflow rates, relative humidity and temperature, were correlated with the diversity and composition of indoor bacterial communities. The relative abundance of bacteria closely related to human pathogens was higher indoors than outdoors, and higher in rooms with lower airflow rates and lower relative humidity. The observed relationship between building design and airborne bacterial diversity suggests that we can manage indoor environments, altering through building design and operation the community of microbial species that potentially colonize the human microbiome during our time indoors."	ENVO:building	TCAG	Architectural design influences the diversity and structure of the built environment microbiome	NA	NA	high-throughput sequencing of the bacterial 16S rRNA gene to quantify relationships between building attributes and airborne bacterial communities at a health-care facility.	FWD:AGAGTTTGATCCTGGCTCAG; REV:TGCTGCCTCCCGTAGGAGT	PMID:22278670	NA	NA	NA	Steven Kembel/Gail Ackermann	U Oregon	GAZ:United States of America	NA	NA	0	Outdoor	NA	NA	C	n	NA	Jessica Green	Green_hospital_air	pyrosequencing	Roof	11.77777778	NA	air metagenome	NA	ENVO:city	FASTA	NA	316.8	NA	NA	NA	green_hospital_air_sloan	Architectural design influences the diversity and structure of the built environment microbiome	655179	U Oregon	NA	NA	NA	1345	"FLX Titanium 27F, 338R"	NA	1130	y	NA	NA	ENVO:air	16S rRNA	NA	Outdoor	V2	Engencore	NA	NA	66.2	NA	percent	45.52	PST	NA	"outside air inlet, 11:30"
O2.489858	CCTTCCTT	CAAGAGTTTGATCCTGGCTCAG	skembel@uoregon.edu	NA	Outdoor air	Outdoor_air	Outdoor Air	source	NA	n	2012	Architectural design influences the diversity and structure of the built environment microbiome	NA	hospital_air_seqs	NA	mimarks-survey	Outdoor	NA	NA	NA	0	NA	NA	9.94	2/27/12	NA	0	NA	NA	NA	NA	NA	607000	jlgreen@uoregon.edu	NA	O2	NA	".1,g"	NA	NA	"Buildings are complex ecosystems that house trillions of microorganisms interacting with each other, with humans and with their environment. Understanding the ecological and evolutionary processes that determine the diversity and composition of the built environment microbiome???the community of microorganisms that live indoors???is important for understanding the relationship between building design, biodiversity and human health. In this study, we used high-throughput sequencing of the bacterial 16S rRNA gene to quantify relationships between building attributes and airborne bacterial communities at a health-care facility. We quantified airborne bacterial community structure and environmental conditions in patient rooms exposed to mechanical or window ventilation and in outdoor air. The phylogenetic diversity of airborne bacterial communities was lower indoors than outdoors, and mechanically ventilated rooms contained less diverse microbial communities than did window-ventilated rooms. Bacterial communities in indoor environments contained many taxa that are absent or rare outdoors, including taxa closely related to potential human pathogens. Building attributes, specifically the source of ventilation air, airflow rates, relative humidity and temperature, were correlated with the diversity and composition of indoor bacterial communities. The relative abundance of bacteria closely related to human pathogens was higher indoors than outdoors, and higher in rooms with lower airflow rates and lower relative humidity. The observed relationship between building design and airborne bacterial diversity suggests that we can manage indoor environments, altering through building design and operation the community of microbial species that potentially colonize the human microbiome during our time indoors."	NA	"none, outside air"	-122.68	y	NA	U Oregon	NA	NA	NA	"Buildings are complex ecosystems that house trillions of microorganisms interacting with each other, with humans and with their environment. Understanding the ecological and evolutionary processes that determine the diversity and composition of the built environment microbiome the community of microorganisms that live indoors is important for understanding the relationship between building design, biodiversity and human health. In this study, we used high-throughput sequencing of the bacterial 16S rRNA gene to quantify relationships between building attributes and airborne bacterial communities at a health-care facility. We quantified airborne bacterial community structure and environmental conditions in patient rooms exposed to mechanical or window ventilation and in outdoor air. The phylogenetic diversity of airborne bacterial communities was lower indoors than outdoors, and mechanically ventilated rooms contained less diverse microbial communities than did window-ventilated rooms. Bacterial communities in indoor environments contained many taxa that are absent or rare outdoors, including taxa closely related to potential human pathogens. Building attributes, specifically the source of ventilation air, airflow rates, relative humidity and temperature, were correlated with the diversity and composition of indoor bacterial communities. The relative abundance of bacteria closely related to human pathogens was higher indoors than outdoors, and higher in rooms with lower airflow rates and lower relative humidity. The observed relationship between building design and airborne bacterial diversity suggests that we can manage indoor environments, altering through building design and operation the community of microbial species that potentially colonize the human microbiome during our time indoors."	ENVO:building	TCAG	Architectural design influences the diversity and structure of the built environment microbiome	NA	NA	high-throughput sequencing of the bacterial 16S rRNA gene to quantify relationships between building attributes and airborne bacterial communities at a health-care facility.	FWD:AGAGTTTGATCCTGGCTCAG; REV:TGCTGCCTCCCGTAGGAGT	PMID:22278670	NA	NA	NA	Steven Kembel/Gail Ackermann	U Oregon	GAZ:United States of America	NA	NA	0	Outdoor	NA	NA	C	n	NA	Jessica Green	Green_hospital_air	pyrosequencing	Roof	12.40740741	NA	air metagenome	NA	ENVO:city	FASTA	NA	322.6666667	NA	NA	NA	green_hospital_air_sloan	Architectural design influences the diversity and structure of the built environment microbiome	655179	U Oregon	NA	NA	NA	1345	"FLX Titanium 27F, 338R"	NA	1300	y	NA	NA	ENVO:air	16S rRNA	NA	Outdoor	V2	Engencore	NA	NA	62.66666667	NA	percent	45.52	PST	NA	"outside air inlet, 13:00"
O3.489856	GGAACCAA	CAAGAGTTTGATCCTGGCTCAG	skembel@uoregon.edu	NA	Outdoor air	Outdoor_air	Outdoor Air	source	NA	n	2012	Architectural design influences the diversity and structure of the built environment microbiome	NA	hospital_air_seqs	NA	mimarks-survey	Outdoor	NA	NA	NA	0	NA	NA	9.94	2/27/12	NA	0	NA	NA	NA	NA	NA	502000	jlgreen@uoregon.edu	NA	O3	NA	".1,g"	NA	NA	"Buildings are complex ecosystems that house trillions of microorganisms interacting with each other, with humans and with their environment. Understanding the ecological and evolutionary processes that determine the diversity and composition of the built environment microbiome???the community of microorganisms that live indoors???is important for understanding the relationship between building design, biodiversity and human health. In this study, we used high-throughput sequencing of the bacterial 16S rRNA gene to quantify relationships between building attributes and airborne bacterial communities at a health-care facility. We quantified airborne bacterial community structure and environmental conditions in patient rooms exposed to mechanical or window ventilation and in outdoor air. The phylogenetic diversity of airborne bacterial communities was lower indoors than outdoors, and mechanically ventilated rooms contained less diverse microbial communities than did window-ventilated rooms. Bacterial communities in indoor environments contained many taxa that are absent or rare outdoors, including taxa closely related to potential human pathogens. Building attributes, specifically the source of ventilation air, airflow rates, relative humidity and temperature, were correlated with the diversity and composition of indoor bacterial communities. The relative abundance of bacteria closely related to human pathogens was higher indoors than outdoors, and higher in rooms with lower airflow rates and lower relative humidity. The observed relationship between building design and airborne bacterial diversity suggests that we can manage indoor environments, altering through building design and operation the community of microbial species that potentially colonize the human microbiome during our time indoors."	NA	"none, outside air"	-122.68	y	NA	U Oregon	NA	NA	NA	"Buildings are complex ecosystems that house trillions of microorganisms interacting with each other, with humans and with their environment. Understanding the ecological and evolutionary processes that determine the diversity and composition of the built environment microbiome the community of microorganisms that live indoors is important for understanding the relationship between building design, biodiversity and human health. In this study, we used high-throughput sequencing of the bacterial 16S rRNA gene to quantify relationships between building attributes and airborne bacterial communities at a health-care facility. We quantified airborne bacterial community structure and environmental conditions in patient rooms exposed to mechanical or window ventilation and in outdoor air. The phylogenetic diversity of airborne bacterial communities was lower indoors than outdoors, and mechanically ventilated rooms contained less diverse microbial communities than did window-ventilated rooms. Bacterial communities in indoor environments contained many taxa that are absent or rare outdoors, including taxa closely related to potential human pathogens. Building attributes, specifically the source of ventilation air, airflow rates, relative humidity and temperature, were correlated with the diversity and composition of indoor bacterial communities. The relative abundance of bacteria closely related to human pathogens was higher indoors than outdoors, and higher in rooms with lower airflow rates and lower relative humidity. The observed relationship between building design and airborne bacterial diversity suggests that we can manage indoor environments, altering through building design and operation the community of microbial species that potentially colonize the human microbiome during our time indoors."	ENVO:building	TCAG	Architectural design influences the diversity and structure of the built environment microbiome	NA	NA	high-throughput sequencing of the bacterial 16S rRNA gene to quantify relationships between building attributes and airborne bacterial communities at a health-care facility.	FWD:AGAGTTTGATCCTGGCTCAG; REV:TGCTGCCTCCCGTAGGAGT	PMID:22278670	NA	NA	NA	Steven Kembel/Gail Ackermann	U Oregon	GAZ:United States of America	NA	NA	0	Outdoor	NA	NA	C	n	NA	Jessica Green	Green_hospital_air	pyrosequencing	Roof	13.33333333	NA	air metagenome	NA	ENVO:city	FASTA	NA	176	NA	NA	NA	green_hospital_air_sloan	Architectural design influences the diversity and structure of the built environment microbiome	655179	U Oregon	NA	NA	NA	1345	"FLX Titanium 27F, 338R"	NA	1600	y	NA	NA	ENVO:air	16S rRNA	NA	Outdoor	V2	Engencore	NA	NA	62.4	NA	percent	45.52	PST	NA	outside air inlet 16:00
O5.489855	TTCCAAGG	CAAGAGTTTGATCCTGGCTCAG	skembel@uoregon.edu	NA	Outdoor air	Outdoor_air	Outdoor Air	source	NA	n	2012	Architectural design influences the diversity and structure of the built environment microbiome	NA	hospital_air_seqs	NA	mimarks-survey	Outdoor	NA	NA	NA	0	NA	NA	9.94	2/28/12 12:20	NA	0	NA	NA	NA	NA	NA	2050000	jlgreen@uoregon.edu	NA	O5	NA	".1,g"	NA	NA	"Buildings are complex ecosystems that house trillions of microorganisms interacting with each other, with humans and with their environment. Understanding the ecological and evolutionary processes that determine the diversity and composition of the built environment microbiome???the community of microorganisms that live indoors???is important for understanding the relationship between building design, biodiversity and human health. In this study, we used high-throughput sequencing of the bacterial 16S rRNA gene to quantify relationships between building attributes and airborne bacterial communities at a health-care facility. We quantified airborne bacterial community structure and environmental conditions in patient rooms exposed to mechanical or window ventilation and in outdoor air. The phylogenetic diversity of airborne bacterial communities was lower indoors than outdoors, and mechanically ventilated rooms contained less diverse microbial communities than did window-ventilated rooms. Bacterial communities in indoor environments contained many taxa that are absent or rare outdoors, including taxa closely related to potential human pathogens. Building attributes, specifically the source of ventilation air, airflow rates, relative humidity and temperature, were correlated with the diversity and composition of indoor bacterial communities. The relative abundance of bacteria closely related to human pathogens was higher indoors than outdoors, and higher in rooms with lower airflow rates and lower relative humidity. The observed relationship between building design and airborne bacterial diversity suggests that we can manage indoor environments, altering through building design and operation the community of microbial species that potentially colonize the human microbiome during our time indoors."	NA	"none, outside air"	-122.68	y	NA	U Oregon	NA	NA	NA	"Buildings are complex ecosystems that house trillions of microorganisms interacting with each other, with humans and with their environment. Understanding the ecological and evolutionary processes that determine the diversity and composition of the built environment microbiome the community of microorganisms that live indoors is important for understanding the relationship between building design, biodiversity and human health. In this study, we used high-throughput sequencing of the bacterial 16S rRNA gene to quantify relationships between building attributes and airborne bacterial communities at a health-care facility. We quantified airborne bacterial community structure and environmental conditions in patient rooms exposed to mechanical or window ventilation and in outdoor air. The phylogenetic diversity of airborne bacterial communities was lower indoors than outdoors, and mechanically ventilated rooms contained less diverse microbial communities than did window-ventilated rooms. Bacterial communities in indoor environments contained many taxa that are absent or rare outdoors, including taxa closely related to potential human pathogens. Building attributes, specifically the source of ventilation air, airflow rates, relative humidity and temperature, were correlated with the diversity and composition of indoor bacterial communities. The relative abundance of bacteria closely related to human pathogens was higher indoors than outdoors, and higher in rooms with lower airflow rates and lower relative humidity. The observed relationship between building design and airborne bacterial diversity suggests that we can manage indoor environments, altering through building design and operation the community of microbial species that potentially colonize the human microbiome during our time indoors."	ENVO:building	TCAG	Architectural design influences the diversity and structure of the built environment microbiome	NA	NA	high-throughput sequencing of the bacterial 16S rRNA gene to quantify relationships between building attributes and airborne bacterial communities at a health-care facility.	FWD:AGAGTTTGATCCTGGCTCAG; REV:TGCTGCCTCCCGTAGGAGT	PMID:22278670	NA	NA	NA	Steven Kembel/Gail Ackermann	U Oregon	GAZ:United States of America	NA	NA	0	Outdoor	NA	NA	C	n	NA	Jessica Green	Green_hospital_air	pyrosequencing	Roof	11.11111111	NA	air metagenome	NA	ENVO:city	FASTA	NA	0	NA	NA	NA	green_hospital_air_sloan	Architectural design influences the diversity and structure of the built environment microbiome	655179	U Oregon	NA	NA	NA	1345	"FLX Titanium 27F, 338R"	NA	1220	y	NA	NA	ENVO:air	16S rRNA	NA	Outdoor	V2	Engencore	NA	NA	78.2	NA	percent	45.52	PST	NA	"outsie air inlet, 12:20 2/28/2012"
F13Knee.140670	GACTGTCATGCA	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	back_of_knees	Human Skin	source	NA	n	11/14/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	back of knees	FKB0RMH	33	mimarks-survey	NA	FMA:Popliteal fossa	NA	NA	0	UBERON:zone of skin of knee	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F1	sample001	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F13Knee	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	F1	popliteal fossae	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F13Knee	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
F14Knee.140634	GAGTATGCAGCC	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	back_of_knees	Human Skin	source	NA	n	11/14/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	back of knees	FKB0RMH	33	mimarks-survey	NA	FMA:Popliteal fossa	NA	NA	0	UBERON:zone of skin of knee	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F1	sample002	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F14Knee	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	F1	popliteal fossae	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F14Knee	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
F23Knee.140816	GCCACTGATAGT	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	back_of_knees	Human Skin	source	NA	n	11/14/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	back of knees	FKB0RMH	31	mimarks-survey	NA	FMA:Popliteal fossa	NA	NA	0	UBERON:zone of skin of knee	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F2	sample003	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F23Knee	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	31	F2	popliteal fossae	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F23Knee	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
F24Knee.140744	GCTAGTCTGAAC	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	back_of_knees	Human Skin	source	NA	n	11/14/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	back of knees	FKB0RMH	31	mimarks-survey	NA	FMA:Popliteal fossa	NA	NA	0	UBERON:zone of skin of knee	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F2	sample004	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F24Knee	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	31	F2	popliteal fossae	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F24Knee	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
F33Knee.140702	GTATGCGCTGTA	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	back_of_knees	Human Skin	source	NA	n	11/14/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	back of knees	FKB0RMH	33	mimarks-survey	NA	FMA:Popliteal fossa	NA	NA	0	UBERON:zone of skin of knee	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F3	sample005	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F33Knee	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	F3	popliteal fossae	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F33Knee	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
F34Knee.140341	GTCTATCGGAGT	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	back_of_knees	Human Skin	source	NA	n	11/14/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	back of knees	FKB0RMH	33	mimarks-survey	NA	FMA:Popliteal fossa	NA	NA	0	UBERON:zone of skin of knee	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F3	sample006	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F34Knee	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	F3	popliteal fossae	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F34Knee	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M13Knee.140288	GAACATGATGAG	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	back_of_knees	Human Skin	source	NA	n	11/14/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	back of knees	FKB0RMH	36	mimarks-survey	NA	FMA:Popliteal fossa	NA	NA	0	UBERON:zone of skin of knee	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M1	sample007	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M13Knee	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	36	M1	popliteal fossae	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M13Knee	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M14Knee.140681	GACCGAGCTATG	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	back_of_knees	Human Skin	source	NA	n	11/14/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	back of knees	FKB0RMH	36	mimarks-survey	NA	FMA:Popliteal fossa	NA	NA	0	UBERON:zone of skin of knee	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M1	sample008	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M14Knee	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	36	M1	popliteal fossae	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M14Knee	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M23Knee.140621	GATCTATCCGAG	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	back_of_knees	Human Skin	source	NA	n	11/14/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	back of knees	FKB0RMH	36	mimarks-survey	NA	FMA:Popliteal fossa	NA	NA	0	UBERON:zone of skin of knee	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M2	sample009	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M23Knee	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	36	M2	popliteal fossae	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M23Knee	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M24Knee.140796	GCACTGAGACGT	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	back_of_knees	Human Skin	source	NA	n	11/14/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	back of knees	FKB0RMH	36	mimarks-survey	NA	FMA:Popliteal fossa	NA	NA	0	UBERON:zone of skin of knee	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M2	sample010	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M24Knee	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	36	M2	popliteal fossae	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M24Knee	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M33Knee.140298	GCTTGCGAGACA	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	back_of_knees	Human Skin	source	NA	n	11/14/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	back of knees	FKB0RMH	33	mimarks-survey	NA	FMA:Popliteal fossa	NA	NA	0	UBERON:zone of skin of knee	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M3	sample011	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M33Knee	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	M3	popliteal fossae	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M33Knee	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M34Knee.140362	GTACGGCATACG	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	back_of_knees	Human Skin	source	NA	n	11/14/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	back of knees	FKB0RMH	33	mimarks-survey	NA	FMA:Popliteal fossa	NA	NA	0	UBERON:zone of skin of knee	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M3	sample012	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M34Knee	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	M3	popliteal fossae	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M34Knee	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M43Knee.140412	GTGGCGATACAC	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	back_of_knees	Human Skin	source	NA	n	11/14/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	back of knees	FKB0RMH	31	mimarks-survey	NA	FMA:Popliteal fossa	NA	NA	0	UBERON:zone of skin of knee	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M4	sample013	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M43Knee	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	31	M4	popliteal fossae	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M43Knee	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M44Knee.140736	TAAGCGCAGCAC	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	back_of_knees	Human Skin	source	NA	n	11/14/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	back of knees	FKB0RMH	31	mimarks-survey	NA	FMA:Popliteal fossa	NA	NA	0	UBERON:zone of skin of knee	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M4	sample014	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M44Knee	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	31	M4	popliteal fossae	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M44Knee	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M53Knee.140318	TACTACATGGTC	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	back_of_knees	Human Skin	source	NA	n	11/14/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	back of knees	FKB0RMH	60	mimarks-survey	NA	FMA:Popliteal fossa	NA	NA	0	UBERON:zone of skin of knee	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M5	sample015	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M53Knee	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	60	M5	popliteal fossae	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M53Knee	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M54Knee.140481	TAGCGACATCTG	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	back_of_knees	Human Skin	source	NA	n	11/14/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	back of knees	FKB0RMH	60	mimarks-survey	NA	FMA:Popliteal fossa	NA	NA	0	UBERON:zone of skin of knee	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M5	sample016	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M54Knee	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	60	M5	popliteal fossae	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M54Knee	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M63Knee.140419	TATCTCGAACTG	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	back_of_knees	Human Skin	source	NA	n	11/14/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	back of knees	FKB0RMH	NA	mimarks-survey	NA	FMA:Popliteal fossa	NA	NA	0	UBERON:zone of skin of knee	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M6	sample017	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M63Knee	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	NA	M6	popliteal fossae	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M63Knee	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M64Knee.140570	TCAGACAGACCG	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	back_of_knees	Human Skin	source	NA	n	11/14/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	back of knees	FKB0RMH	NA	mimarks-survey	NA	FMA:Popliteal fossa	NA	NA	0	UBERON:zone of skin of knee	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M6	sample018	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M64Knee	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	NA	M6	popliteal fossae	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M64Knee	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
F11Nose.140325	AGCAGCACTTGT	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	external_nose	Human Skin	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	external nose	FFLHOYS	33	mimarks-survey	NA	FMA:External nose	NA	NA	0	UBERON:nose	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F1	sample019	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F11Nose	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	F1	external nose	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F11Nose	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
F12Nose.140365	AGTGTTCGATCG	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	external_nose	Human Skin	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	external nose	FFLHOYS	33	mimarks-survey	NA	FMA:External nose	NA	NA	0	UBERON:nose	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F1	sample020	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F12Nose	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	F1	external nose	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F12Nose	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
F21Nose.140573	ATGACTCATTCG	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	external_nose	Human Skin	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	external nose	FFLHOYS	31	mimarks-survey	NA	FMA:External nose	NA	NA	0	UBERON:nose	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F2	sample021	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F21Nose	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	31	F2	external nose	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F21Nose	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
F22Nose.140758	CACGGACTATAC	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	external_nose	Human Skin	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	external nose	FFLHOYS	31	mimarks-survey	NA	FMA:External nose	NA	NA	0	UBERON:nose	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F2	sample022	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F22Nose	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	31	F2	external nose	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F22Nose	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
F31Nose.140465	GCAGCACGTTGA	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	external_nose	Human Skin	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	external nose	FFO92CG	33	mimarks-survey	NA	FMA:External nose	NA	NA	0	UBERON:nose	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F3	sample023	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F31Nose	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	F3	external nose	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFO92CG	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F31Nose	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
F32Nose.140543	GCTGATGAGCTG	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	external_nose	Human Skin	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	external nose	FFO92CG	33	mimarks-survey	NA	FMA:External nose	NA	NA	0	UBERON:nose	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F3	sample024	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F32Nose	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	F3	external nose	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFO92CG	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F32Nose	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M11Nose.140428	ACAGTGCTTCAT	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	external_nose	Human Skin	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	external nose	FFLHOYS	36	mimarks-survey	NA	FMA:External nose	NA	NA	0	UBERON:nose	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M1	sample025	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M11Nose	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	36	M1	external nose	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M11Nose	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M12Nose.140763	ACTACGTGTGGT	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	external_nose	Human Skin	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	external nose	FFLHOYS	36	mimarks-survey	NA	FMA:External nose	NA	NA	0	UBERON:nose	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M1	sample026	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M12Nose	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	36	M1	external nose	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M12Nose	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M21Nose.140682	CATATCGCAGTT	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	external_nose	Human Skin	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	external nose	FFO92CG	36	mimarks-survey	NA	FMA:External nose	NA	NA	0	UBERON:nose	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M2	sample027	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M21Nose	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	36	M2	external nose	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFO92CG	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M21Nose	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M22Nose.140594	CGAGTTGTAGCG	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	external_nose	Human Skin	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	external nose	FFO92CG	36	mimarks-survey	NA	FMA:External nose	NA	NA	0	UBERON:nose	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M2	sample028	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M22Nose	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	36	M2	external nose	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFO92CG	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M22Nose	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M31Nose.140415	CGTGATCTCTCC	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	external_nose	Human Skin	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	external nose	FFLHOYS	33	mimarks-survey	NA	FMA:External nose	NA	NA	0	UBERON:nose	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M3	sample029	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M31Nose	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	M3	external nose	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M31Nose	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M32Nose.140531	CTCGATTAGATC	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	external_nose	Human Skin	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	external nose	FFLHOYS	33	mimarks-survey	NA	FMA:External nose	NA	NA	0	UBERON:nose	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M3	sample030	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M32Nose	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	M3	external nose	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M32Nose	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M41Nose.140779	GAATGATGAGTG	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	external_nose	Human Skin	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	external nose	FFO92CG	31	mimarks-survey	NA	FMA:External nose	NA	NA	0	UBERON:nose	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M4	sample031	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M41Nose	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	31	M4	external nose	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFO92CG	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M41Nose	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M42Nose.140308	GAGCTGGCTGAT	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	external_nose	Human Skin	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	external nose	FFO92CG	31	mimarks-survey	NA	FMA:External nose	NA	NA	0	UBERON:nose	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M4	sample032	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M42Nose	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	31	M4	external nose	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFO92CG	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M42Nose	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
F11Frhd.140508	AGAGTCCTGAGC	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	forehead	Human Skin	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	forehead	FFLHOYS	33	mimarks-survey	NA	FMA:Forehead	NA	NA	0	UBERON:zone of skin of head	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F1	sample033	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F11Frhd	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	F1	forehead	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F11Frhd	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
F12Frhd.140469	AGTCTACTCTGA	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	forehead	Human Skin	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	forehead	FFLHOYS	33	mimarks-survey	NA	FMA:Forehead	NA	NA	0	UBERON:zone of skin of head	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F1	sample034	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F12Frhd	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	F1	forehead	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F12Frhd	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
F13Frhd.140649	GACTGATCATCT	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	forehead	Human Skin	source	NA	n	11/14/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	forehead	FKB0RMH	33	mimarks-survey	NA	FMA:Forehead	NA	NA	0	UBERON:zone of skin of head	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F1	sample035	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F13Frhd	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	F1	forehead	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F13Frhd	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
F14Frhd.140741	GAGGCTCATCAT	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	forehead	Human Skin	source	NA	n	11/14/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	forehead	FKB0RMH	33	mimarks-survey	NA	FMA:Forehead	NA	NA	0	UBERON:zone of skin of head	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F1	sample036	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F14Frhd	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	F1	forehead	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F14Frhd	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
F21Frhd.140623	ATCGTACAACTC	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	forehead	Human Skin	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	forehead	FFLHOYS	31	mimarks-survey	NA	FMA:Forehead	NA	NA	0	UBERON:zone of skin of head	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F2	sample037	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F21Frhd	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	31	F2	forehead	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F21Frhd	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
F22Frhd.140836	CAAGTGAGAGAG	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	forehead	Human Skin	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	forehead	FFLHOYS	31	mimarks-survey	NA	FMA:Forehead	NA	NA	0	UBERON:zone of skin of head	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F2	sample038	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F22Frhd	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	31	F2	forehead	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F22Frhd	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
F23Frhd.140622	GCATGTGCATGT	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	forehead	Human Skin	source	NA	n	11/14/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	forehead	FKB0RMH	31	mimarks-survey	NA	FMA:Forehead	NA	NA	0	UBERON:zone of skin of head	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F2	sample039	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F23Frhd	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	31	F2	forehead	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F23Frhd	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
F24Frhd.140858	GCTAAGAGAGTA	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	forehead	Human Skin	source	NA	n	11/14/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	forehead	FKB0RMH	31	mimarks-survey	NA	FMA:Forehead	NA	NA	0	UBERON:zone of skin of head	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F2	sample040	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F24Frhd	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	31	F2	forehead	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F24Frhd	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
F31Frhd.140832	GATGTGAGCGCT	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	forehead	Human Skin	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	forehead	FFO92CG	33	mimarks-survey	NA	FMA:Forehead	NA	NA	0	UBERON:zone of skin of head	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F3	sample041	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F31Frhd	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	F3	forehead	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFO92CG	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F31Frhd	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
F32Frhd.140444	GCTAAGAGAGTA	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	forehead	Human Skin	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	forehead	FFO92CG	33	mimarks-survey	NA	FMA:Forehead	NA	NA	0	UBERON:zone of skin of head	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F3	sample042	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F32Frhd	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	F3	forehead	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFO92CG	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F32Frhd	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
F33Frhd.140757	GTATCCATGCGA	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	forehead	Human Skin	source	NA	n	11/14/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	forehead	FKB0RMH	33	mimarks-survey	NA	FMA:Forehead	NA	NA	0	UBERON:zone of skin of head	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F3	sample043	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F33Frhd	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	F3	forehead	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F33Frhd	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
F34Frhd.140568	GTCGTGTGTCAA	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	forehead	Human Skin	source	NA	n	11/14/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	forehead	FKB0RMH	33	mimarks-survey	NA	FMA:Forehead	NA	NA	0	UBERON:zone of skin of head	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F3	sample044	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F34Frhd	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	F3	forehead	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F34Frhd	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M11Frhd.140548	ACACGGTGTCTA	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	forehead	Human Skin	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	forehead	FFLHOYS	36	mimarks-survey	NA	FMA:Forehead	NA	NA	0	UBERON:zone of skin of head	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M1	sample045	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M11Frhd	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	36	M1	forehead	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M11Frhd	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M12Frhd.140766	ACGGTGAGTGTC	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	forehead	Human Skin	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	forehead	FFLHOYS	36	mimarks-survey	NA	FMA:Forehead	NA	NA	0	UBERON:zone of skin of head	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M1	sample046	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M12Frhd	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	36	M1	forehead	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M12Frhd	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M13Frhd.140349	CTTGATGCGTAT	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	forehead	Human Skin	source	NA	n	11/14/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	forehead	FKB0RMH	36	mimarks-survey	NA	FMA:Forehead	NA	NA	0	UBERON:zone of skin of head	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M1	sample047	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M13Frhd	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	36	M1	forehead	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M13Frhd	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M14Frhd.140825	GACATCGGCTAT	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	forehead	Human Skin	source	NA	n	11/14/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	forehead	FKB0RMH	36	mimarks-survey	NA	FMA:Forehead	NA	NA	0	UBERON:zone of skin of head	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M1	sample048	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M14Frhd	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	36	M1	forehead	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M14Frhd	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M21Frhd.140641	CAGTGATCCTAG	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	forehead	Human Skin	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	forehead	FFO92CG	36	mimarks-survey	NA	FMA:Forehead	NA	NA	0	UBERON:zone of skin of head	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M2	sample049	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M21Frhd	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	36	M2	forehead	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFO92CG	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M21Frhd	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M22Frhd.140287	CGACAGCTGACA	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	forehead	Human Skin	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	forehead	FFO92CG	36	mimarks-survey	NA	FMA:Forehead	NA	NA	0	UBERON:zone of skin of head	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M2	sample050	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M22Frhd	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	36	M2	forehead	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFO92CG	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M22Frhd	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M23Frhd.140404	GATCGCAGGTGT	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	forehead	Human Skin	source	NA	n	11/14/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	forehead	FKB0RMH	36	mimarks-survey	NA	FMA:Forehead	NA	NA	0	UBERON:zone of skin of head	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M2	sample051	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M23Frhd	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	36	M2	forehead	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M23Frhd	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M24Frhd.140442	GCACGACAACAC	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	forehead	Human Skin	source	NA	n	11/14/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	forehead	FKB0RMH	36	mimarks-survey	NA	FMA:Forehead	NA	NA	0	UBERON:zone of skin of head	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M2	sample052	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M24Frhd	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	36	M2	forehead	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M24Frhd	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M31Frhd.140336	CGTATCTGCGAA	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	forehead	Human Skin	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	forehead	FFLHOYS	33	mimarks-survey	NA	FMA:Forehead	NA	NA	0	UBERON:zone of skin of head	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M3	sample053	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M31Frhd	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	M3	forehead	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M31Frhd	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M32Frhd.140776	CTATGCTTGATG	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	forehead	Human Skin	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	forehead	FFLHOYS	33	mimarks-survey	NA	FMA:Forehead	NA	NA	0	UBERON:zone of skin of head	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M3	sample054	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M32Frhd	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	M3	forehead	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M32Frhd	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M33Frhd.140580	GCTTACATCGAG	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	forehead	Human Skin	source	NA	n	11/14/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	forehead	FKB0RMH	33	mimarks-survey	NA	FMA:Forehead	NA	NA	0	UBERON:zone of skin of head	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M3	sample055	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M33Frhd	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	M3	forehead	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M33Frhd	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M34Frhd.140638	GGTGCGTGTATG	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	forehead	Human Skin	source	NA	n	11/14/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	forehead	FKB0RMH	33	mimarks-survey	NA	FMA:Forehead	NA	NA	0	UBERON:zone of skin of head	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M3	sample056	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M34Frhd	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	M3	forehead	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M34Frhd	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M41Frhd.140449	CTTGATGCGTAT	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	forehead	Human Skin	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	forehead	FFO92CG	31	mimarks-survey	NA	FMA:Forehead	NA	NA	0	UBERON:zone of skin of head	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M4	sample057	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M41Frhd	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	31	M4	forehead	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFO92CG	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M41Frhd	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M42Frhd.140770	GAGAATACGTGA	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	forehead	Human Skin	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	forehead	FFO92CG	31	mimarks-survey	NA	FMA:Forehead	NA	NA	0	UBERON:zone of skin of head	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M4	sample058	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M42Frhd	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	31	M4	forehead	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFO92CG	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M42Frhd	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M43Frhd.140486	GTGCAATCGACG	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	forehead	Human Skin	source	NA	n	11/14/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	forehead	FKB0RMH	31	mimarks-survey	NA	FMA:Forehead	NA	NA	0	UBERON:zone of skin of head	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M4	sample059	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M43Frhd	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	31	M4	forehead	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M43Frhd	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M44Frhd.140730	TAACAGTCGCTG	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	forehead	Human Skin	source	NA	n	11/14/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	forehead	FKB0RMH	31	mimarks-survey	NA	FMA:Forehead	NA	NA	0	UBERON:zone of skin of head	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M4	sample060	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M44Frhd	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	31	M4	forehead	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M44Frhd	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M53Frhd.140768	TACGTGTACGTG	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	forehead	Human Skin	source	NA	n	11/14/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	forehead	FKB0RMH	60	mimarks-survey	NA	FMA:Forehead	NA	NA	0	UBERON:zone of skin of head	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M5	sample061	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M53Frhd	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	60	M5	forehead	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M53Frhd	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M54Frhd.140695	TAGCATCGTGGT	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	forehead	Human Skin	source	NA	n	11/14/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	forehead	FKB0RMH	60	mimarks-survey	NA	FMA:Forehead	NA	NA	0	UBERON:zone of skin of head	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M5	sample062	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M54Frhd	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	60	M5	forehead	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M54Frhd	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M63Frhd.140286	TATCAGGTGTGC	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	forehead	Human Skin	source	NA	n	11/14/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	forehead	FKB0RMH	NA	mimarks-survey	NA	FMA:Forehead	NA	NA	0	UBERON:zone of skin of head	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M6	sample063	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M63Frhd	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	NA	M6	forehead	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M63Frhd	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M64Frhd.140834	TCACTGGCAGTA	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	forehead	Human Skin	source	NA	n	11/14/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	forehead	FKB0RMH	NA	mimarks-survey	NA	FMA:Forehead	NA	NA	0	UBERON:zone of skin of head	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M6	sample064	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M64Frhd	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	NA	M6	forehead	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M64Frhd	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M11Glns.140664	ACACTAGATCCG	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	glans_penis	Human Skin	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	glans penis	FFLHOYS	36	mimarks-survey	NA	FMA:Glans penis	NA	NA	0	UBERON:glans penis	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M1	sample065	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M11Glns	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	36	M1	glans penis	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M11Glns	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M12Glns.140683	ACGTACTCAGTG	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	glans_penis	Human Skin	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	glans penis	FFLHOYS	36	mimarks-survey	NA	FMA:Glans penis	NA	NA	0	UBERON:glans penis	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M1	sample066	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M12Glns	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	36	M1	glans penis	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M12Glns	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M21Glns.140311	CAGTGCATATGC	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	glans_penis	Human Skin	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	glans penis	FFO92CG	36	mimarks-survey	NA	FMA:Glans penis	NA	NA	0	UBERON:glans penis	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M2	sample067	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M21Glns	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	36	M2	glans penis	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFO92CG	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M21Glns	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M22Glns.140314	CGACATGCTATT	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	glans_penis	Human Skin	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	glans penis	FFO92CG	36	mimarks-survey	NA	FMA:Glans penis	NA	NA	0	UBERON:glans penis	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M2	sample068	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M22Glns	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	36	M2	glans penis	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFO92CG	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M22Glns	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M31Glns.140477	CGTATGCTGTAT	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	glans_penis	Human Skin	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	glans penis	FFLHOYS	33	mimarks-survey	NA	FMA:Glans penis	NA	NA	0	UBERON:glans penis	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M3	sample069	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M31Glns	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	M3	glans penis	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M31Glns	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M32Glns.140441	CTCAATGACTCA	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	glans_penis	Human Skin	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	glans penis	FFLHOYS	33	mimarks-survey	NA	FMA:Glans penis	NA	NA	0	UBERON:glans penis	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M3	sample070	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M32Glns	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	M3	glans penis	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M32Glns	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M42Glns.140719	GAGACAGCTTGC	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	glans_penis	Human Skin	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	glans penis	FFO92CG	31	mimarks-survey	NA	FMA:Glans penis	NA	NA	0	UBERON:glans penis	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M4	sample072	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M42Glns	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	31	M4	glans penis	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFO92CG	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M42Glns	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
F11Hair.140612	AGATACACGCGC	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	hair_on_head	Human Skin	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	hair on head	FFLHOYS	33	mimarks-survey	NA	FMA:Head hair	NA	NA	0	UBERON:hair	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F1	sample073	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:hair	UBERON:hair	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F11Hair	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	F1	hair on head	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human hair metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	646099	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F11Hair	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human hair metagenome
F12Hair.140386	AGTCTCGCATAT	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	hair_on_head	Human Skin	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	hair on head	FFLHOYS	33	mimarks-survey	NA	FMA:Head hair	NA	NA	0	UBERON:hair	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F1	sample074	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:hair	UBERON:hair	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F12Hair	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	F1	hair on head	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human hair metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	646099	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F12Hair	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human hair metagenome
F21Hair.140524	ATCTACTACACG	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	hair_on_head	Human Skin	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	hair on head	FFLHOYS	31	mimarks-survey	NA	FMA:Head hair	NA	NA	0	UBERON:hair	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F2	sample075	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:hair	UBERON:hair	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F21Hair	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	31	F2	hair on head	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human hair metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	646099	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F21Hair	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human hair metagenome
F22Hair.140512	CACACGTGAGCA	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	hair_on_head	Human Skin	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	hair on head	FFLHOYS	31	mimarks-survey	NA	FMA:Head hair	NA	NA	0	UBERON:hair	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F2	sample076	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:hair	UBERON:hair	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F22Hair	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	31	F2	hair on head	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human hair metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	646099	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F22Hair	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human hair metagenome
F31Hair.140753	GATTAGCACTCT	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	hair_on_head	Human Skin	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	hair on head	FFO92CG	33	mimarks-survey	NA	FMA:Head hair	NA	NA	0	UBERON:hair	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F3	sample077	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:hair	UBERON:hair	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F31Hair	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	F3	hair on head	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human hair metagenome	NA	ENVO:human-associated habitat	FLX	FFO92CG	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	646099	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F31Hair	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human hair metagenome
F32Hair.140769	GCTAGATGCCAG	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	hair_on_head	Human Skin	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	hair on head	FFO92CG	33	mimarks-survey	NA	FMA:Head hair	NA	NA	0	UBERON:hair	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F3	sample078	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:hair	UBERON:hair	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F32Hair	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	F3	hair on head	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human hair metagenome	NA	ENVO:human-associated habitat	FLX	FFO92CG	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	646099	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F32Hair	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human hair metagenome
M11Hair.140577	ACACTGTTCATG	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	hair_on_head	Human Skin	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	hair on head	FFLHOYS	36	mimarks-survey	NA	FMA:Head hair	NA	NA	0	UBERON:hair	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M1	sample079	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:hair	UBERON:hair	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M11Hair	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	36	M1	hair on head	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human hair metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	646099	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M11Hair	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human hair metagenome
M12Hair.140590	ACGTCTGTAGCA	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	hair_on_head	Human Skin	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	hair on head	FFLHOYS	36	mimarks-survey	NA	FMA:Head hair	NA	NA	0	UBERON:hair	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M1	sample080	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:hair	UBERON:hair	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M12Hair	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	36	M1	hair on head	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human hair metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	646099	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M12Hair	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human hair metagenome
M21Hair.140504	CAGTGTCAGGAC	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	hair_on_head	Human Skin	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	hair on head	FFO92CG	36	mimarks-survey	NA	FMA:Head hair	NA	NA	0	UBERON:hair	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M2	sample081	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:hair	UBERON:hair	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M21Hair	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	36	M2	hair on head	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human hair metagenome	NA	ENVO:human-associated habitat	FLX	FFO92CG	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	646099	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M21Hair	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human hair metagenome
M22Hair.140456	CGACTTATGTGT	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	hair_on_head	Human Skin	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	hair on head	FFO92CG	36	mimarks-survey	NA	FMA:Head hair	NA	NA	0	UBERON:hair	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M2	sample082	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:hair	UBERON:hair	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M22Hair	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	36	M2	hair on head	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human hair metagenome	NA	ENVO:human-associated habitat	FLX	FFO92CG	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	646099	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M22Hair	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human hair metagenome
M31Hair.140713	CGTCAACGATGT	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	hair_on_head	Human Skin	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	hair on head	FFLHOYS	33	mimarks-survey	NA	FMA:Head hair	NA	NA	0	UBERON:hair	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M3	sample083	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:hair	UBERON:hair	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M31Hair	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	M3	hair on head	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human hair metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	646099	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M31Hair	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human hair metagenome
M32Hair.140734	CTCAGTATGCAG	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	hair_on_head	Human Skin	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	hair on head	FFLHOYS	33	mimarks-survey	NA	FMA:Head hair	NA	NA	0	UBERON:hair	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M3	sample084	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:hair	UBERON:hair	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M32Hair	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	M3	hair on head	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human hair metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	646099	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M32Hair	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human hair metagenome
M41Hair.140413	GAACATGATGAG	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	hair_on_head	Human Skin	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	hair on head	FFO92CG	31	mimarks-survey	NA	FMA:Head hair	NA	NA	0	UBERON:hair	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M4	sample085	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:hair	UBERON:hair	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M41Hair	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	31	M4	hair on head	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human hair metagenome	NA	ENVO:human-associated habitat	FLX	FFO92CG	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	646099	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M41Hair	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human hair metagenome
M42Hair.140700	GAGAGAATGATC	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	hair_on_head	Human Skin	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	hair on head	FFO92CG	31	mimarks-survey	NA	FMA:Head hair	NA	NA	0	UBERON:hair	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M4	sample086	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:hair	UBERON:hair	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M42Hair	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	31	M4	hair on head	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human hair metagenome	NA	ENVO:human-associated habitat	FLX	FFO92CG	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	646099	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M42Hair	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human hair metagenome
F11Labi.140687	AGATGTTCTGCT	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	labia_minora	Human Vagina	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	labia minora	FFLHOYS	33	mimarks-survey	NA	FMA:Frenulum of labia minora	NA	NA	0	UBERON:labia minora	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F1	sample087	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:mucus	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F11Labi	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	F1	labia minora	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F11Labi	ENVO:mucus	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
F12Labi.140407	AGTTAGTGCGTC	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	labia_minora	Human Vagina	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	labia minora	FFLHOYS	33	mimarks-survey	NA	FMA:Frenulum of labia minora	NA	NA	0	UBERON:labia minora	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F1	sample088	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:mucus	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F12Labi	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	F1	labia minora	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F12Labi	ENVO:mucus	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
F21Labi.140606	ATCTGGTGCTAT	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	labia_minora	Human Vagina	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	labia minora	FFLHOYS	31	mimarks-survey	NA	FMA:Frenulum of labia minora	NA	NA	0	UBERON:labia minora	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F2	sample089	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:mucus	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F21Labi	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	31	F2	labia minora	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F21Labi	ENVO:mucus	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
F22Labi.140560	CACATCTAACAC	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	labia_minora	Human Vagina	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	labia minora	FFLHOYS	31	mimarks-survey	NA	FMA:Frenulum of labia minora	NA	NA	0	UBERON:labia minora	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F2	sample090	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:mucus	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F22Labi	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	31	F2	labia minora	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F22Labi	ENVO:mucus	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
F31Labi.140819	GCACGACAACAC	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	labia_minora	Human Vagina	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	labia minora	FFO92CG	33	mimarks-survey	NA	FMA:Frenulum of labia minora	NA	NA	0	UBERON:labia minora	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F3	sample091	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:mucus	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F31Labi	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	F3	labia minora	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFO92CG	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F31Labi	ENVO:mucus	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
F32Labi.140727	GCTATTCGACAT	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	labia_minora	Human Vagina	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	labia minora	FFO92CG	33	mimarks-survey	NA	FMA:Frenulum of labia minora	NA	NA	0	UBERON:labia minora	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F3	sample092	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:mucus	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F32Labi	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	F3	labia minora	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFO92CG	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F32Labi	ENVO:mucus	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
F11Aptl.140780	ACTGTCGAAGCT	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	left_armpit	Human Skin	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	left armpit	FFLHOYS	33	mimarks-survey	NA	FMA:Left axilla	NA	NA	0	UBERON:skin of arm	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F1	sample093	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F11Aptl	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	F1	left axilla	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F11Aptl	ENVO:sweat	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
F12Aptl.140591	AGCTGACTAGTC	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	left_armpit	Human Skin	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	left armpit	FFLHOYS	33	mimarks-survey	NA	FMA:Left axilla	NA	NA	0	UBERON:skin of arm	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F1	sample094	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F12Aptl	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	F1	left axilla	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F12Aptl	ENVO:sweat	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
F21Aptl.140686	ATATCGCTACTG	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	left_armpit	Human Skin	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	left armpit	FFLHOYS	31	mimarks-survey	NA	FMA:Left axilla	NA	NA	0	UBERON:skin of arm	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F2	sample095	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F21Aptl	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	31	F2	left axilla	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F21Aptl	ENVO:sweat	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
F22Aptl.140370	ATGTACGGCGAC	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	left_armpit	Human Skin	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	left armpit	FFLHOYS	31	mimarks-survey	NA	FMA:Left axilla	NA	NA	0	UBERON:skin of arm	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F2	sample096	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F22Aptl	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	31	F2	left axilla	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F22Aptl	ENVO:sweat	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
F31Aptl.140409	GATCAGAAGATG	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	left_armpit	Human Skin	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	left armpit	FFO92CG	33	mimarks-survey	NA	FMA:Left axilla	NA	NA	0	UBERON:skin of arm	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F3	sample097	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F31Aptl	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	F3	left axilla	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFO92CG	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F31Aptl	ENVO:sweat	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
F32Aptl.140820	GCCACTGATAGT	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	left_armpit	Human Skin	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	left armpit	FFO92CG	33	mimarks-survey	NA	FMA:Left axilla	NA	NA	0	UBERON:skin of arm	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F3	sample098	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F32Aptl	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	F3	left axilla	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFO92CG	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F32Aptl	ENVO:sweat	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M11Aptl.140830	AACGCACGCTAG	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	left_armpit	Human Skin	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	left armpit	FFLHOYS	36	mimarks-survey	NA	FMA:Left axilla	NA	NA	0	UBERON:skin of arm	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M1	sample099	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M11Aptl	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	36	M1	left axilla	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M11Aptl	ENVO:sweat	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M12Aptl.140844	ACCTGTCTCTCT	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	left_armpit	Human Skin	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	left armpit	FFLHOYS	36	mimarks-survey	NA	FMA:Left axilla	NA	NA	0	UBERON:skin of arm	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M1	sample100	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M12Aptl	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	36	M1	left axilla	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M12Aptl	ENVO:sweat	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M21Aptl.140554	CAGATACACTTC	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	left_armpit	Human Skin	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	left armpit	FFO92CG	36	mimarks-survey	NA	FMA:Left axilla	NA	NA	0	UBERON:skin of arm	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M2	sample101	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M21Aptl	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	36	M2	left axilla	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFO92CG	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M21Aptl	ENVO:sweat	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M31Aptl.140756	CGCGATAGCAGT	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	left_armpit	Human Skin	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	left armpit	FFLHOYS	33	mimarks-survey	NA	FMA:Left axilla	NA	NA	0	UBERON:skin of arm	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M3	sample103	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M31Aptl	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	M3	left axilla	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M31Aptl	ENVO:sweat	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M32Aptl.140296	CTACTACAGGTG	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	left_armpit	Human Skin	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	left armpit	FFLHOYS	33	mimarks-survey	NA	FMA:Left axilla	NA	NA	0	UBERON:skin of arm	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M3	sample104	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M32Aptl	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	M3	left axilla	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M32Aptl	ENVO:sweat	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M41Aptl.140841	CTGCTGCGAAGA	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	left_armpit	Human Skin	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	left armpit	FFO92CG	31	mimarks-survey	NA	FMA:Left axilla	NA	NA	0	UBERON:skin of arm	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M4	sample105	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M41Aptl	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	31	M4	left axilla	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFO92CG	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M41Aptl	ENVO:sweat	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M42Aptl.140537	GACGCTAGTTCA	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	left_armpit	Human Skin	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	left armpit	FFO92CG	31	mimarks-survey	NA	FMA:Left axilla	NA	NA	0	UBERON:skin of arm	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M4	sample106	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M42Aptl	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	31	M4	left axilla	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFO92CG	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M42Aptl	ENVO:sweat	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
F11Knel.140331	ACTTGTAGCAGC	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	left_back_of_knees	Human Skin	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	left back of knees	FFLHOYS	33	mimarks-survey	NA	FMA:Left popliteal fossa	NA	NA	0	UBERON:zone of skin of knee	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F1	sample107	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F11Knel	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	F1	left popliteal fossa	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F11Knel	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
F12Knel.140738	AGGACGCACTGT	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	left_back_of_knees	Human Skin	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	left back of knees	FFLHOYS	33	mimarks-survey	NA	FMA:Left popliteal fossa	NA	NA	0	UBERON:zone of skin of knee	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F1	sample108	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F12Knel	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	F1	left popliteal fossa	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F12Knel	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
F21Knel.140372	ATCACGTAGCGG	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	left_back_of_knees	Human Skin	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	left back of knees	FFLHOYS	31	mimarks-survey	NA	FMA:Left popliteal fossa	NA	NA	0	UBERON:zone of skin of knee	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F2	sample109	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F21Knel	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	31	F2	left popliteal fossa	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F21Knel	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
F22Knel.140720	ATGTGCACGACT	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	left_back_of_knees	Human Skin	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	left back of knees	FFLHOYS	31	mimarks-survey	NA	FMA:Left popliteal fossa	NA	NA	0	UBERON:zone of skin of knee	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F2	sample110	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F22Knel	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	31	F2	left popliteal fossa	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F22Knel	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
F31Knel.140379	GATCGCAGGTGT	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	left_back_of_knees	Human Skin	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	left back of knees	FFO92CG	33	mimarks-survey	NA	FMA:Left popliteal fossa	NA	NA	0	UBERON:zone of skin of knee	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F3	sample111	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F31Knel	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	F3	left popliteal fossa	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFO92CG	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F31Knel	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
F32Knel.140324	GCCTATACTACA	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	left_back_of_knees	Human Skin	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	left back of knees	FFO92CG	33	mimarks-survey	NA	FMA:Left popliteal fossa	NA	NA	0	UBERON:zone of skin of knee	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F3	sample112	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F32Knel	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	F3	left popliteal fossa	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFO92CG	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F32Knel	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M11Knel.140676	AACTGTGCGTAC	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	left_back_of_knees	Human Skin	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	left back of knees	FFLHOYS	36	mimarks-survey	NA	FMA:Left popliteal fossa	NA	NA	0	UBERON:zone of skin of knee	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M1	sample113	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M11Knel	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	36	M1	left popliteal fossa	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M11Knel	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M12Knel.140275	ACGAGTGCTATC	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	left_back_of_knees	Human Skin	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	left back of knees	FFLHOYS	36	mimarks-survey	NA	FMA:Left popliteal fossa	NA	NA	0	UBERON:zone of skin of knee	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M1	sample114	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M12Knel	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	36	M1	left popliteal fossa	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M12Knel	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M21Knel.140490	CAGCACTAAGCG	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	left_back_of_knees	Human Skin	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	left back of knees	FFO92CG	36	mimarks-survey	NA	FMA:Left popliteal fossa	NA	NA	0	UBERON:zone of skin of knee	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M2	sample115	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M21Knel	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	36	M2	left popliteal fossa	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFO92CG	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M21Knel	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M31Knel.140873	CGCTAGAACGCA	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	left_back_of_knees	Human Skin	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	left back of knees	FFLHOYS	33	mimarks-survey	NA	FMA:Left popliteal fossa	NA	NA	0	UBERON:zone of skin of knee	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M3	sample117	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M31Knel	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	M3	left popliteal fossa	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M31Knel	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M41Knel.140855	CTGGAGCATGAC	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	left_back_of_knees	Human Skin	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	left back of knees	FFO92CG	31	mimarks-survey	NA	FMA:Left popliteal fossa	NA	NA	0	UBERON:zone of skin of knee	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M4	sample119	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M41Knel	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	31	M4	left popliteal fossa	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFO92CG	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M41Knel	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M42Knel.140699	GACTAACGTCAC	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	left_back_of_knees	Human Skin	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	left back of knees	FFO92CG	31	mimarks-survey	NA	FMA:Left popliteal fossa	NA	NA	0	UBERON:zone of skin of knee	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M4	sample120	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M42Knel	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	31	M4	left popliteal fossa	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFO92CG	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M42Knel	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
F11Forl.140437	AGAGCAAGAGCA	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	left_forearm	Human Skin	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	left forearm	FFLHOYS	33	mimarks-survey	NA	FMA:Surface of left arm	NA	NA	0	UBERON:skin of forearm	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F1	sample121	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F11Forl	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	F1	left volar forearm	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F11Forl	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
F21Forl.140295	ATCGCGGACGAT	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	left_forearm	Human Skin	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	left forearm	FFLHOYS	31	mimarks-survey	NA	FMA:Surface of left arm	NA	NA	0	UBERON:skin of forearm	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F2	sample123	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F21Forl	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	31	F2	left volar forearm	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F21Forl	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
F22Forl.140279	CAACTCATCGTA	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	left_forearm	Human Skin	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	left forearm	FFLHOYS	31	mimarks-survey	NA	FMA:Surface of left arm	NA	NA	0	UBERON:skin of forearm	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F2	sample124	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F22Forl	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	31	F2	left volar forearm	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F22Forl	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
F31Forl.140333	GATGCATGACGC	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	left_forearm	Human Skin	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	left forearm	FFO92CG	33	mimarks-survey	NA	FMA:Surface of left arm	NA	NA	0	UBERON:skin of forearm	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F3	sample125	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F31Forl	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	F3	left volar forearm	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFO92CG	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F31Forl	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
F32Forl.140724	GCGTATCTTGAT	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	left_forearm	Human Skin	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	left forearm	FFO92CG	33	mimarks-survey	NA	FMA:Surface of left arm	NA	NA	0	UBERON:skin of forearm	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F3	sample126	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F32Forl	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	F3	left volar forearm	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFO92CG	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F32Forl	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M11Forl.140337	ACACATGTCTAC	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	left_forearm	Human Skin	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	left forearm	FFLHOYS	36	mimarks-survey	NA	FMA:Surface of left arm	NA	NA	0	UBERON:skin of forearm	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M1	sample127	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M11Forl	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	36	M1	left volar forearm	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M11Forl	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M12Forl.140812	ACGCTCATGGAT	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	left_forearm	Human Skin	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	left forearm	FFLHOYS	36	mimarks-survey	NA	FMA:Surface of left arm	NA	NA	0	UBERON:skin of forearm	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M1	sample128	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M12Forl	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	36	M1	left volar forearm	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M12Forl	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M21Forl.140771	CAGTCACTAACG	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	left_forearm	Human Skin	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	left forearm	FFO92CG	36	mimarks-survey	NA	FMA:Surface of left arm	NA	NA	0	UBERON:skin of forearm	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M2	sample129	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M21Forl	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	36	M2	left volar forearm	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFO92CG	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M21Forl	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M22Forl.140587	CGAAGACTGCTG	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	left_forearm	Human Skin	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	left forearm	FFO92CG	36	mimarks-survey	NA	FMA:Surface of left arm	NA	NA	0	UBERON:skin of forearm	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M2	sample130	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M22Forl	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	36	M2	left volar forearm	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFO92CG	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M22Forl	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M31Forl.140292	CGTACTAGACTG	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	left_forearm	Human Skin	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	left forearm	FFLHOYS	33	mimarks-survey	NA	FMA:Surface of left arm	NA	NA	0	UBERON:skin of forearm	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M3	sample131	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M31Forl	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	M3	left volar forearm	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M31Forl	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M32Forl.140592	CTATCAGTGTAC	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	left_forearm	Human Skin	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	left forearm	FFLHOYS	33	mimarks-survey	NA	FMA:Surface of left arm	NA	NA	0	UBERON:skin of forearm	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M3	sample132	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M32Forl	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	M3	left volar forearm	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M32Forl	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M41Forl.140496	CTGTTCGTAGAG	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	left_forearm	Human Skin	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	left forearm	FFO92CG	31	mimarks-survey	NA	FMA:Surface of left arm	NA	NA	0	UBERON:skin of forearm	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M4	sample133	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M41Forl	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	31	M4	left volar forearm	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFO92CG	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M41Forl	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M42Forl.140600	GACTGTCATGCA	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	left_forearm	Human Skin	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	left forearm	FFO92CG	31	mimarks-survey	NA	FMA:Surface of left arm	NA	NA	0	UBERON:skin of forearm	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M4	sample134	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M42Forl	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	31	M4	left volar forearm	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFO92CG	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M42Forl	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
F11Indl.140283	AGATCGGCTCGA	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	left_index_finger	Human Skin	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	left index finger	FFLHOYS	33	mimarks-survey	NA	FMA:Left index finger	NA	NA	0	UBERON:skin of finger	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F1	sample135	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F11Indl	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	F1	left palmar index finger	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F11Indl	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
F12Indl.140710	AGTGAGAGAAGC	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	left_index_finger	Human Skin	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	left index finger	FFLHOYS	33	mimarks-survey	NA	FMA:Left index finger	NA	NA	0	UBERON:skin of finger	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F1	sample136	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F12Indl	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	F1	left palmar index finger	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F12Indl	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
F21Indl.140446	ATCTCTGGCATA	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	left_index_finger	Human Skin	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	left index finger	FFLHOYS	31	mimarks-survey	NA	FMA:Left index finger	NA	NA	0	UBERON:skin of finger	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F2	sample137	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F21Indl	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	31	F2	left palmar index finger	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F21Indl	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
F22Indl.140644	CACAGCTCGAAT	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	left_index_finger	Human Skin	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	left index finger	FFLHOYS	31	mimarks-survey	NA	FMA:Left index finger	NA	NA	0	UBERON:skin of finger	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F2	sample138	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F22Indl	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	31	F2	left palmar index finger	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F22Indl	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
F31Indl.140679	GCAATAGCTGCT	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	left_index_finger	Human Skin	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	left index finger	FFO92CG	33	mimarks-survey	NA	FMA:Left index finger	NA	NA	0	UBERON:skin of finger	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F3	sample139	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F31Indl	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	F3	left palmar index finger	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFO92CG	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F31Indl	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
F32Indl.140607	GCTAGTCTGAAC	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	left_index_finger	Human Skin	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	left index finger	FFO92CG	33	mimarks-survey	NA	FMA:Left index finger	NA	NA	0	UBERON:skin of finger	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F3	sample140	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F32Indl	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	F3	left palmar index finger	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFO92CG	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F32Indl	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M11Indl.140749	ACAGACCACTCA	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	left_index_finger	Human Skin	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	left index finger	FFLHOYS	36	mimarks-survey	NA	FMA:Left index finger	NA	NA	0	UBERON:skin of finger	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M1	sample141	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M11Indl	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	36	M1	left palmar index finger	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M11Indl	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M12Indl.140438	ACGTGAGAGAAT	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	left_index_finger	Human Skin	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	left index finger	FFLHOYS	36	mimarks-survey	NA	FMA:Left index finger	NA	NA	0	UBERON:skin of finger	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M1	sample142	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M12Indl	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	36	M1	left palmar index finger	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M12Indl	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M21Indl.140576	CATACCAGTAGC	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	left_index_finger	Human Skin	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	left index finger	FFO92CG	36	mimarks-survey	NA	FMA:Left index finger	NA	NA	0	UBERON:skin of finger	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M2	sample143	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M21Indl	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	36	M2	left palmar index finger	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFO92CG	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M21Indl	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M22Indl.140813	CGAGAGTTACGC	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	left_index_finger	Human Skin	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	left index finger	FFO92CG	36	mimarks-survey	NA	FMA:Left index finger	NA	NA	0	UBERON:skin of finger	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M2	sample144	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M22Indl	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	36	M2	left palmar index finger	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFO92CG	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M22Indl	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M31Indl.140291	CGTCACGACTAA	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	left_index_finger	Human Skin	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	left index finger	FFLHOYS	33	mimarks-survey	NA	FMA:Left index finger	NA	NA	0	UBERON:skin of finger	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M3	sample145	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M31Indl	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	M3	left palmar index finger	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M31Indl	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M32Indl.140814	CTCATGTACAGT	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	left_index_finger	Human Skin	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	left index finger	FFLHOYS	33	mimarks-survey	NA	FMA:Left index finger	NA	NA	0	UBERON:skin of finger	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M3	sample146	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M32Indl	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	M3	left palmar index finger	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M32Indl	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M41Indl.140854	GAACTGTATCTC	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	left_index_finger	Human Skin	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	left index finger	FFO92CG	31	mimarks-survey	NA	FMA:Left index finger	NA	NA	0	UBERON:skin of finger	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M4	sample147	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M41Indl	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	31	M4	left palmar index finger	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFO92CG	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M41Indl	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M42Indl.140869	GAGAGCTCTACG	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	left_index_finger	Human Skin	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	left index finger	FFO92CG	31	mimarks-survey	NA	FMA:Left index finger	NA	NA	0	UBERON:skin of finger	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M4	sample148	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M42Indl	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	31	M4	left palmar index finger	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFO92CG	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M42Indl	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
F11Nstl.140539	AGCAGTCGCGAT	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	left_nostril	Human Skin	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	left nostril	FFLHOYS	33	mimarks-survey	NA	FMA:Nares	NA	NA	0	UBERON:nostril	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F1	sample149	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:nostril	UBERON:mucus	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F11Nstl	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	F1	left naris	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human nose metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	646099	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F11Nstl	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human nose metagenome
F12Nstl.140831	AGTTCAGACGCT	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	left_nostril	Human Skin	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	left nostril	FFLHOYS	33	mimarks-survey	NA	FMA:Nares	NA	NA	0	UBERON:nostril	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F1	sample150	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:nostril	UBERON:mucus	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F12Nstl	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	F1	left naris	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human nose metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	646099	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F12Nstl	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human nose metagenome
F21Nstl.140383	ATGAGACTCCAC	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	left_nostril	Human Skin	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	left nostril	FFLHOYS	31	mimarks-survey	NA	FMA:Nares	NA	NA	0	UBERON:nostril	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F2	sample151	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:nostril	UBERON:mucus	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F21Nstl	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	31	F2	left naris	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human nose metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	646099	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F21Nstl	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human nose metagenome
F22Nstl.140534	CACGTCGATGGA	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	left_nostril	Human Skin	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	left nostril	FFLHOYS	31	mimarks-survey	NA	FMA:Nares	NA	NA	0	UBERON:nostril	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F2	sample152	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:nostril	UBERON:mucus	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F22Nstl	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	31	F2	left naris	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human nose metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	646099	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F22Nstl	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human nose metagenome
F31Nstl.140789	GCAGCCGAGTAT	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	left_nostril	Human Skin	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	left nostril	FFO92CG	33	mimarks-survey	NA	FMA:Nares	NA	NA	0	UBERON:nostril	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F3	sample153	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:nostril	UBERON:mucus	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F31Nstl	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	F3	left naris	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human nose metagenome	NA	ENVO:human-associated habitat	FLX	FFO92CG	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	646099	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F31Nstl	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human nose metagenome
F32Nstl.140804	GCTGCTGCAATA	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	left_nostril	Human Skin	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	left nostril	FFO92CG	33	mimarks-survey	NA	FMA:Nares	NA	NA	0	UBERON:nostril	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F3	sample154	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:nostril	UBERON:mucus	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F32Nstl	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	F3	left naris	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human nose metagenome	NA	ENVO:human-associated habitat	FLX	FFO92CG	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	646099	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F32Nstl	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human nose metagenome
M11Nstl.140810	ACAGTTGCGCGA	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	left_nostril	Human Skin	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	left nostril	FFLHOYS	36	mimarks-survey	NA	FMA:Nares	NA	NA	0	UBERON:nostril	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M1	sample155	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:nostril	UBERON:mucus	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M11Nstl	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	36	M1	left naris	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human nose metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	646099	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M11Nstl	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human nose metagenome
M12Nstl.140399	ACTAGCTCCATA	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	left_nostril	Human Skin	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	left nostril	FFLHOYS	36	mimarks-survey	NA	FMA:Nares	NA	NA	0	UBERON:nostril	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M1	sample156	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:nostril	UBERON:mucus	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M12Nstl	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	36	M1	left naris	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human nose metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	646099	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M12Nstl	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human nose metagenome
M21Nstl.140369	CATCAGCGTGTA	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	left_nostril	Human Skin	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	left nostril	FFO92CG	36	mimarks-survey	NA	FMA:Nares	NA	NA	0	UBERON:nostril	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M2	sample157	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:nostril	UBERON:mucus	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M21Nstl	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	36	M2	left naris	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human nose metagenome	NA	ENVO:human-associated habitat	FLX	FFO92CG	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	646099	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M21Nstl	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human nose metagenome
M22Nstl.140860	CGATAGATCTTC	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	left_nostril	Human Skin	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	left nostril	FFO92CG	36	mimarks-survey	NA	FMA:Nares	NA	NA	0	UBERON:nostril	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M2	sample158	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:nostril	UBERON:mucus	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M22Nstl	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	36	M2	left naris	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human nose metagenome	NA	ENVO:human-associated habitat	FLX	FFO92CG	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	646099	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M22Nstl	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human nose metagenome
M31Nstl.140392	CGTGCATTATCA	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	left_nostril	Human Skin	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	left nostril	FFLHOYS	33	mimarks-survey	NA	FMA:Nares	NA	NA	0	UBERON:nostril	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M3	sample159	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:nostril	UBERON:mucus	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M31Nstl	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	M3	left naris	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human nose metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	646099	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M31Nstl	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human nose metagenome
M32Nstl.140853	CTCGCACATATA	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	left_nostril	Human Skin	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	left nostril	FFLHOYS	33	mimarks-survey	NA	FMA:Nares	NA	NA	0	UBERON:nostril	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M3	sample160	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:nostril	UBERON:mucus	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M32Nstl	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	M3	left naris	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human nose metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	646099	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M32Nstl	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human nose metagenome
M41Nstl.140739	GACACTCGAATC	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	left_nostril	Human Skin	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	left nostril	FFO92CG	31	mimarks-survey	NA	FMA:Nares	NA	NA	0	UBERON:nostril	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M4	sample161	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:nostril	UBERON:mucus	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M41Nstl	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	31	M4	left naris	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human nose metagenome	NA	ENVO:human-associated habitat	FLX	FFO92CG	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	646099	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M41Nstl	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human nose metagenome
M42Nstl.140363	GAGGCTCATCAT	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	left_nostril	Human Skin	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	left nostril	FFO92CG	31	mimarks-survey	NA	FMA:Nares	NA	NA	0	UBERON:nostril	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M4	sample162	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:nostril	UBERON:mucus	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M42Nstl	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	31	M4	left naris	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human nose metagenome	NA	ENVO:human-associated habitat	FLX	FFO92CG	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	646099	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M42Nstl	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human nose metagenome
F11Pinl.140315	AGCGAGCTATCT	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	left_outer_ear	Human Skin	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	left outer ear	FFLHOYS	33	mimarks-survey	NA	FMA:Surface of left pinna	NA	NA	0	UBERON:zone of skin of outer ear	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F1	sample163	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F11Pinl	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	F1	left lateral pinna	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F11Pinl	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
F12Pinl.140479	ATACAGAGCTCC	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	left_outer_ear	Human Skin	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	left outer ear	FFLHOYS	33	mimarks-survey	NA	FMA:Surface of left pinna	NA	NA	0	UBERON:zone of skin of outer ear	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F1	sample164	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F12Pinl	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	F1	left lateral pinna	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F12Pinl	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
F22Pinl.140717	CACTCTGATTAG	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	left_outer_ear	Human Skin	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	left outer ear	FFLHOYS	31	mimarks-survey	NA	FMA:Surface of left pinna	NA	NA	0	UBERON:zone of skin of outer ear	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F2	sample165	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F22Pinl	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	31	F2	left lateral pinna	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F22Pinl	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
F31Pinl.140395	GCAGTTCATATC	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	left_outer_ear	Human Skin	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	left outer ear	FFO92CG	33	mimarks-survey	NA	FMA:Surface of left pinna	NA	NA	0	UBERON:zone of skin of outer ear	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F3	sample166	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F31Pinl	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	F3	left lateral pinna	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFO92CG	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F31Pinl	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
F32Pinl.140578	GCTTACATCGAG	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	left_outer_ear	Human Skin	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	left outer ear	FFO92CG	33	mimarks-survey	NA	FMA:Surface of left pinna	NA	NA	0	UBERON:zone of skin of outer ear	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F3	sample167	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F32Pinl	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	F3	left lateral pinna	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFO92CG	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F32Pinl	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M11Pinl.140316	ACATTCAGCGCA	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	left_outer_ear	Human Skin	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	left outer ear	FFLHOYS	36	mimarks-survey	NA	FMA:Surface of left pinna	NA	NA	0	UBERON:zone of skin of outer ear	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M1	sample168	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M11Pinl	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	36	M1	left lateral pinna	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M11Pinl	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M12Pinl.140722	ACTCGATTCGAT	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	left_outer_ear	Human Skin	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	left outer ear	FFLHOYS	36	mimarks-survey	NA	FMA:Surface of left pinna	NA	NA	0	UBERON:zone of skin of outer ear	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M1	sample169	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M12Pinl	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	36	M1	left lateral pinna	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M12Pinl	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M21Pinl.140313	CATGAGTGCTAC	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	left_outer_ear	Human Skin	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	left outer ear	FFO92CG	36	mimarks-survey	NA	FMA:Surface of left pinna	NA	NA	0	UBERON:zone of skin of outer ear	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M2	sample170	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M21Pinl	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	36	M2	left lateral pinna	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFO92CG	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M21Pinl	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M22Pinl.140526	CGATGTCGTCAA	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	left_outer_ear	Human Skin	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	left outer ear	FFO92CG	36	mimarks-survey	NA	FMA:Surface of left pinna	NA	NA	0	UBERON:zone of skin of outer ear	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M2	sample171	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M22Pinl	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	36	M2	left lateral pinna	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFO92CG	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M22Pinl	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M31Pinl.140703	CGTTATGTACAC	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	left_outer_ear	Human Skin	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	left outer ear	FFLHOYS	33	mimarks-survey	NA	FMA:Surface of left pinna	NA	NA	0	UBERON:zone of skin of outer ear	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M3	sample172	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M31Pinl	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	M3	left lateral pinna	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M31Pinl	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M32Pinl.140666	CTCTGCTAGCCT	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	left_outer_ear	Human Skin	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	left outer ear	FFLHOYS	33	mimarks-survey	NA	FMA:Surface of left pinna	NA	NA	0	UBERON:zone of skin of outer ear	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M3	sample173	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M32Pinl	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	M3	left lateral pinna	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M32Pinl	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M41Pinl.140284	GACATCGGCTAT	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	left_outer_ear	Human Skin	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	left outer ear	FFO92CG	31	mimarks-survey	NA	FMA:Surface of left pinna	NA	NA	0	UBERON:zone of skin of outer ear	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M4	sample174	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M41Pinl	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	31	M4	left lateral pinna	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFO92CG	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M41Pinl	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M42Pinl.140354	GAGTGAGTACAA	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	left_outer_ear	Human Skin	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	left outer ear	FFO92CG	31	mimarks-survey	NA	FMA:Surface of left pinna	NA	NA	0	UBERON:zone of skin of outer ear	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M4	sample175	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M42Pinl	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	31	M4	left lateral pinna	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFO92CG	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M42Pinl	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
F11Ewxl.140795	AGACCGTCAGAC	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	left_outer_ear_canal_earwax	Human Skin	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	left outer ear canal/earwax	FFLHOYS	33	mimarks-survey	NA	FMA:Left external auditory canal	NA	NA	0	UBERON:ear canal	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F1	sample176	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:external auditory canal	UBERON:cerumen	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F11Ewxl	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	F1	left external auditory canal	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	646099	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F11Ewxl	ENVO:ear wax	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human metagenome
F12Ewxl.140466	AGGTGTGATCGC	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	left_outer_ear_canal_earwax	Human Skin	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	left outer ear canal/earwax	FFLHOYS	33	mimarks-survey	NA	FMA:Left external auditory canal	NA	NA	0	UBERON:ear canal	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F1	sample177	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:external auditory canal	UBERON:cerumen	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F12Ewxl	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	F1	left external auditory canal	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	646099	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F12Ewxl	ENVO:ear wax	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human metagenome
F21Ewxl.140312	ATCAGGCGTGTG	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	left_outer_ear_canal_earwax	Human Skin	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	left outer ear canal/earwax	FFLHOYS	31	mimarks-survey	NA	FMA:Left external auditory canal	NA	NA	0	UBERON:ear canal	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F2	sample178	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:external auditory canal	UBERON:cerumen	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F21Ewxl	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	31	F2	left external auditory canal	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	646099	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F21Ewxl	ENVO:ear wax	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human metagenome
F22Ewxl.140693	ATTATCGTGCAC	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	left_outer_ear_canal_earwax	Human Skin	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	left outer ear canal/earwax	FFLHOYS	31	mimarks-survey	NA	FMA:Left external auditory canal	NA	NA	0	UBERON:ear canal	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F2	sample179	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:external auditory canal	UBERON:cerumen	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F22Ewxl	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	31	F2	left external auditory canal	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	646099	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F22Ewxl	ENVO:ear wax	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human metagenome
F32Ewxl.140542	GCGAGATCCAGT	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	left_outer_ear_canal_earwax	Human Skin	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	left outer ear canal/earwax	FFO92CG	33	mimarks-survey	NA	FMA:Left external auditory canal	NA	NA	0	UBERON:ear canal	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F3	sample180	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:external auditory canal	UBERON:cerumen	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F32Ewxl	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	F3	left external auditory canal	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human metagenome	NA	ENVO:human-associated habitat	FLX	FFO92CG	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	646099	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F32Ewxl	ENVO:ear wax	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human metagenome
M11Ewxl.140424	AAGCTGCAGTCG	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	left_outer_ear_canal_earwax	Human Skin	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	left outer ear canal/earwax	FFLHOYS	36	mimarks-survey	NA	FMA:Left external auditory canal	NA	NA	0	UBERON:ear canal	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M1	sample181	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:external auditory canal	UBERON:cerumen	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M11Ewxl	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	36	M1	left external auditory canal	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	646099	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M11Ewxl	ENVO:ear wax	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human metagenome
M12Ewxl.140401	ACGCAACTGCTA	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	left_outer_ear_canal_earwax	Human Skin	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	left outer ear canal/earwax	FFLHOYS	36	mimarks-survey	NA	FMA:Left external auditory canal	NA	NA	0	UBERON:ear canal	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M1	sample182	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:external auditory canal	UBERON:cerumen	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M12Ewxl	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	36	M1	left external auditory canal	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	646099	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M12Ewxl	ENVO:ear wax	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human metagenome
M21Ewxl.140436	CAGCGGTGACAT	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	left_outer_ear_canal_earwax	Human Skin	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	left outer ear canal/earwax	FFO92CG	36	mimarks-survey	NA	FMA:Left external auditory canal	NA	NA	0	UBERON:ear canal	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M2	sample183	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:external auditory canal	UBERON:cerumen	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M21Ewxl	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	36	M2	left external auditory canal	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human metagenome	NA	ENVO:human-associated habitat	FLX	FFO92CG	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	646099	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M21Ewxl	ENVO:ear wax	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human metagenome
M22Ewxl.140750	CCGACTGAGATG	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	left_outer_ear_canal_earwax	Human Skin	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	left outer ear canal/earwax	FFO92CG	36	mimarks-survey	NA	FMA:Left external auditory canal	NA	NA	0	UBERON:ear canal	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M2	sample184	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:external auditory canal	UBERON:cerumen	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M22Ewxl	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	36	M2	left external auditory canal	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human metagenome	NA	ENVO:human-associated habitat	FLX	FFO92CG	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	646099	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M22Ewxl	ENVO:ear wax	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human metagenome
M31Ewxl.140375	CGGAGTGTCTAT	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	left_outer_ear_canal_earwax	Human Skin	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	left outer ear canal/earwax	FFLHOYS	33	mimarks-survey	NA	FMA:Left external auditory canal	NA	NA	0	UBERON:ear canal	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M3	sample185	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:external auditory canal	UBERON:cerumen	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M31Ewxl	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	M3	left external auditory canal	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	646099	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M31Ewxl	ENVO:ear wax	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human metagenome
M32Ewxl.140473	CTAGCGAACATC	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	left_outer_ear_canal_earwax	Human Skin	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	left outer ear canal/earwax	FFLHOYS	33	mimarks-survey	NA	FMA:Left external auditory canal	NA	NA	0	UBERON:ear canal	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M3	sample186	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:external auditory canal	UBERON:cerumen	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M32Ewxl	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	M3	left external auditory canal	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	646099	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M32Ewxl	ENVO:ear wax	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human metagenome
M41Ewxl.140482	CTGTATCGTATG	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	left_outer_ear_canal_earwax	Human Skin	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	left outer ear canal/earwax	FFO92CG	31	mimarks-survey	NA	FMA:Left external auditory canal	NA	NA	0	UBERON:ear canal	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M4	sample187	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:external auditory canal	UBERON:cerumen	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M41Ewxl	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	31	M4	left external auditory canal	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human metagenome	NA	ENVO:human-associated habitat	FLX	FFO92CG	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	646099	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M41Ewxl	ENVO:ear wax	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human metagenome
M42Ewxl.140553	GACTCACTCAAT	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	left_outer_ear_canal_earwax	Human Skin	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	left outer ear canal/earwax	FFO92CG	31	mimarks-survey	NA	FMA:Left external auditory canal	NA	NA	0	UBERON:ear canal	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M4	sample188	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:external auditory canal	UBERON:cerumen	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M42Ewxl	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	31	M4	left external auditory canal	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human metagenome	NA	ENVO:human-associated habitat	FLX	FFO92CG	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	646099	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M42Ewxl	ENVO:ear wax	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human metagenome
F11Plml.140803	AGCCATACTGAC	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	left_palm	Human Skin	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	left palm	FFLHOYS	33	mimarks-survey	NA	FMA:Left palm	NA	NA	0	UBERON:zone of skin of hand	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F1	sample189	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F11Plml	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	F1	left palm	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F11Plml	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
F12Plml.140300	ATAATCTCGTCG	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	left_palm	Human Skin	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	left palm	FFLHOYS	33	mimarks-survey	NA	FMA:Left palm	NA	NA	0	UBERON:zone of skin of hand	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F1	sample190	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F12Plml	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	F1	left palm	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F12Plml	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
F13Plml.140601	GACTCGAATCGT	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	left_palm	Human Skin	source	NA	n	11/14/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	left palm	FKB0RMH	33	mimarks-survey	NA	FMA:Left palm	NA	NA	0	UBERON:zone of skin of hand	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F1	sample191	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F13Plml	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	F1	left palm	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F13Plml	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
F14Plml.140525	GAGCTGGCTGAT	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	left_palm	Human Skin	source	NA	n	11/14/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	left palm	FKB0RMH	33	mimarks-survey	NA	FMA:Left palm	NA	NA	0	UBERON:zone of skin of hand	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F1	sample192	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F14Plml	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	F1	left palm	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F14Plml	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
F21Plml.140704	ATGCACTGGCGA	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	left_palm	Human Skin	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	left palm	FFLHOYS	31	mimarks-survey	NA	FMA:Left palm	NA	NA	0	UBERON:zone of skin of hand	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F2	sample193	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F21Plml	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	31	F2	left palm	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F21Plml	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
F22Plml.140347	CACTACTGTTGA	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	left_palm	Human Skin	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	left palm	FFLHOYS	31	mimarks-survey	NA	FMA:Left palm	NA	NA	0	UBERON:zone of skin of hand	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F2	sample194	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F22Plml	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	31	F2	left palm	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F22Plml	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
F23Plml.140585	GCATCGTCAACA	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	left_palm	Human Skin	source	NA	n	11/14/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	left palm	FKB0RMH	31	mimarks-survey	NA	FMA:Left palm	NA	NA	0	UBERON:zone of skin of hand	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F2	sample195	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F23Plml	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	31	F2	left palm	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F23Plml	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
F24Plml.140452	GCGTTACACACA	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	left_palm	Human Skin	source	NA	n	11/14/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	left palm	FKB0RMH	31	mimarks-survey	NA	FMA:Left palm	NA	NA	0	UBERON:zone of skin of hand	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F2	sample196	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F24Plml	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	31	F2	left palm	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F24Plml	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
F31Plml.140322	GCAGGCAGTACT	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	left_palm	Human Skin	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	left palm	FFO92CG	33	mimarks-survey	NA	FMA:Left palm	NA	NA	0	UBERON:zone of skin of hand	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F3	sample197	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F31Plml	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	F3	left palm	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFO92CG	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F31Plml	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
F32Plml.140737	GCTGTAGTATGC	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	left_palm	Human Skin	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	left palm	FFO92CG	33	mimarks-survey	NA	FMA:Left palm	NA	NA	0	UBERON:zone of skin of hand	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F3	sample198	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F32Plml	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	F3	left palm	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFO92CG	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F32Plml	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
F33Plml.140709	GTATATCCGCAG	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	left_palm	Human Skin	source	NA	n	11/14/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	left palm	FKB0RMH	33	mimarks-survey	NA	FMA:Left palm	NA	NA	0	UBERON:zone of skin of hand	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F3	sample199	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F33Plml	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	F3	left palm	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F33Plml	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
F34Plml.140846	GTCGTAGCCAGA	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	left_palm	Human Skin	source	NA	n	11/14/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	left palm	FKB0RMH	33	mimarks-survey	NA	FMA:Left palm	NA	NA	0	UBERON:zone of skin of hand	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F3	sample200	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F34Plml	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	F3	left palm	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F34Plml	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M11Plml.140620	ACATGATCGTTC	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	left_palm	Human Skin	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	left palm	FFLHOYS	36	mimarks-survey	NA	FMA:Left palm	NA	NA	0	UBERON:zone of skin of hand	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M1	sample201	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M11Plml	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	36	M1	left palm	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M11Plml	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M12Plml.140817	ACTCACGGTATG	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	left_palm	Human Skin	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	left palm	FFLHOYS	36	mimarks-survey	NA	FMA:Left palm	NA	NA	0	UBERON:zone of skin of hand	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M1	sample202	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M12Plml	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	36	M1	left palm	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M12Plml	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M13Plml.140472	CTTAGCACATCA	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	left_palm	Human Skin	source	NA	n	11/14/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	left palm	FKB0RMH	36	mimarks-survey	NA	FMA:Left palm	NA	NA	0	UBERON:zone of skin of hand	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M1	sample203	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M13Plml	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	36	M1	left palm	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M13Plml	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M14Plml.140544	GACAGTTACTGC	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	left_palm	Human Skin	source	NA	n	11/14/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	left palm	FKB0RMH	36	mimarks-survey	NA	FMA:Left palm	NA	NA	0	UBERON:zone of skin of hand	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M1	sample204	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M14Plml	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	36	M1	left palm	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M14Plml	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M21Plml.140352	CATCGTATCAAC	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	left_palm	Human Skin	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	left palm	FFO92CG	36	mimarks-survey	NA	FMA:Left palm	NA	NA	0	UBERON:zone of skin of hand	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M2	sample205	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M21Plml	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	36	M2	left palm	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFO92CG	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M21Plml	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M22Plml.140625	CGATCGAGTGTT	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	left_palm	Human Skin	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	left palm	FFO92CG	36	mimarks-survey	NA	FMA:Left palm	NA	NA	0	UBERON:zone of skin of hand	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M2	sample206	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M22Plml	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	36	M2	left palm	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFO92CG	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M22Plml	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M23Plml.140503	GATCCGACACTA	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	left_palm	Human Skin	source	NA	n	11/14/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	left palm	FKB0RMH	36	mimarks-survey	NA	FMA:Left palm	NA	NA	0	UBERON:zone of skin of hand	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M2	sample207	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M23Plml	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	36	M2	left palm	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M23Plml	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M24Plml.140651	GCACATCGAGCA	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	left_palm	Human Skin	source	NA	n	11/14/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	left palm	FKB0RMH	36	mimarks-survey	NA	FMA:Left palm	NA	NA	0	UBERON:zone of skin of hand	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M2	sample208	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M24Plml	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	36	M2	left palm	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M24Plml	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M31Plml.140743	CGTGTGATCAGG	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	left_palm	Human Skin	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	left palm	FFLHOYS	33	mimarks-survey	NA	FMA:Left palm	NA	NA	0	UBERON:zone of skin of hand	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M3	sample209	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M31Plml	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	M3	left palm	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M31Plml	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M32Plml.140507	CTCTCTACCTGT	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	left_palm	Human Skin	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	left palm	FFLHOYS	33	mimarks-survey	NA	FMA:Left palm	NA	NA	0	UBERON:zone of skin of hand	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M3	sample210	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M32Plml	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	M3	left palm	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M32Plml	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M33Plml.140822	GCTGTGTAGGAC	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	left_palm	Human Skin	source	NA	n	11/14/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	left palm	FKB0RMH	33	mimarks-survey	NA	FMA:Left palm	NA	NA	0	UBERON:zone of skin of hand	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M3	sample211	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M33Plml	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	M3	left palm	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M33Plml	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M34Plml.140373	GGTCGTAGCGTA	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	left_palm	Human Skin	source	NA	n	11/14/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	left palm	FKB0RMH	33	mimarks-survey	NA	FMA:Left palm	NA	NA	0	UBERON:zone of skin of hand	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M3	sample212	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M34Plml	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	M3	left palm	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M34Plml	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M41Plml.140691	GACAGGAGATAG	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	left_palm	Human Skin	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	left palm	FFO92CG	31	mimarks-survey	NA	FMA:Left palm	NA	NA	0	UBERON:zone of skin of hand	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M4	sample213	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M41Plml	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	31	M4	left palm	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFO92CG	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M41Plml	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M42Plml.140513	GAGTATGCAGCC	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	left_palm	Human Skin	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	left palm	FFO92CG	31	mimarks-survey	NA	FMA:Left palm	NA	NA	0	UBERON:zone of skin of hand	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M4	sample214	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M42Plml	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	31	M4	left palm	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFO92CG	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M42Plml	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M43Plml.140405	GTGATAGTGCCG	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	left_palm	Human Skin	source	NA	n	11/14/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	left palm	FKB0RMH	31	mimarks-survey	NA	FMA:Left palm	NA	NA	0	UBERON:zone of skin of hand	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M4	sample215	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M43Plml	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	31	M4	left palm	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M43Plml	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M44Plml.140614	GTTGTATACTCG	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	left_palm	Human Skin	source	NA	n	11/14/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	left palm	FKB0RMH	31	mimarks-survey	NA	FMA:Left palm	NA	NA	0	UBERON:zone of skin of hand	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M4	sample216	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M44Plml	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	31	M4	left palm	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M44Plml	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M53Plml.140660	TACGGTATGTCT	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	left_palm	Human Skin	source	NA	n	11/14/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	left palm	FKB0RMH	60	mimarks-survey	NA	FMA:Left palm	NA	NA	0	UBERON:zone of skin of hand	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M5	sample217	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M53Plml	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	60	M5	left palm	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M53Plml	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M54Plml.140636	TAGCACACCTAT	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	left_palm	Human Skin	source	NA	n	11/14/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	left palm	FKB0RMH	60	mimarks-survey	NA	FMA:Left palm	NA	NA	0	UBERON:zone of skin of hand	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M5	sample218	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M54Plml	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	60	M5	left palm	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M54Plml	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M63Plml.140689	TATACGCGCATT	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	left_palm	Human Skin	source	NA	n	11/14/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	left palm	FKB0RMH	NA	mimarks-survey	NA	FMA:Left palm	NA	NA	0	UBERON:zone of skin of hand	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M6	sample219	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M63Plml	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	NA	M6	left palm	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M63Plml	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M64Plml.140628	TCACTATGGTCA	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	left_palm	Human Skin	source	NA	n	11/14/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	left palm	FKB0RMH	NA	mimarks-survey	NA	FMA:Left palm	NA	NA	0	UBERON:zone of skin of hand	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M6	sample220	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M64Plml	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	NA	M6	left palm	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M64Plml	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
F11Fotl.140558	AGCGTAGGTCGT	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	left_sole_of_foot	Human Skin	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	left sole of foot	FFLHOYS	33	mimarks-survey	NA	FMA:Left foot surface	NA	NA	0	UBERON:zone of skin of foot	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F1	sample221	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F11Fotl	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	F1	left plantar foot	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F11Fotl	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
F12Fotl.140294	ATACTATTGCGC	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	left_sole_of_foot	Human Skin	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	left sole of foot	FFLHOYS	33	mimarks-survey	NA	FMA:Left foot surface	NA	NA	0	UBERON:zone of skin of foot	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F1	sample222	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F12Fotl	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	F1	left plantar foot	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F12Fotl	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
F13Fotl.140488	GAGAATACGTGA	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	left_sole_of_foot	Human Skin	source	NA	n	11/14/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	left sole of foot	FKB0RMH	33	mimarks-survey	NA	FMA:Left foot surface	NA	NA	0	UBERON:zone of skin of foot	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F1	sample223	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F13Fotl	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	F1	left plantar foot	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F13Fotl	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
F14Fotl.140552	GAGTGAGTACAA	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	left_sole_of_foot	Human Skin	source	NA	n	11/14/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	left sole of foot	FKB0RMH	33	mimarks-survey	NA	FMA:Left foot surface	NA	NA	0	UBERON:zone of skin of foot	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F1	sample224	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F14Fotl	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	F1	left plantar foot	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F14Fotl	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
F21Fotl.140348	ATGGATACGCTC	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	left_sole_of_foot	Human Skin	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	left sole of foot	FFLHOYS	31	mimarks-survey	NA	FMA:Left foot surface	NA	NA	0	UBERON:zone of skin of foot	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F2	sample225	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F21Fotl	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	31	F2	left plantar foot	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F21Fotl	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
F22Fotl.140421	CACTGTAGGACG	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	left_sole_of_foot	Human Skin	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	left sole of foot	FFLHOYS	31	mimarks-survey	NA	FMA:Left foot surface	NA	NA	0	UBERON:zone of skin of foot	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F2	sample226	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F22Fotl	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	31	F2	left plantar foot	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F22Fotl	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
F23Fotl.140863	GCCTATACTACA	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	left_sole_of_foot	Human Skin	source	NA	n	11/14/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	left sole of foot	FKB0RMH	31	mimarks-survey	NA	FMA:Left foot surface	NA	NA	0	UBERON:zone of skin of foot	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F2	sample227	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F23Fotl	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	31	F2	left plantar foot	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F23Fotl	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
F24Fotl.140630	GCTATTCGACAT	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	left_sole_of_foot	Human Skin	source	NA	n	11/14/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	left sole of foot	FKB0RMH	31	mimarks-survey	NA	FMA:Left foot surface	NA	NA	0	UBERON:zone of skin of foot	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F2	sample228	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F24Fotl	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	31	F2	left plantar foot	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F24Fotl	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
F31Fotl.140715	GCATATAGTCTC	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	left_sole_of_foot	Human Skin	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	left sole of foot	FFO92CG	33	mimarks-survey	NA	FMA:Left foot surface	NA	NA	0	UBERON:zone of skin of foot	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F3	sample229	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F31Fotl	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	F3	left plantar foot	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFO92CG	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F31Fotl	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
F32Fotl.140455	GCTTGCGAGACA	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	left_sole_of_foot	Human Skin	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	left sole of foot	FFO92CG	33	mimarks-survey	NA	FMA:Left foot surface	NA	NA	0	UBERON:zone of skin of foot	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F3	sample230	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F32Fotl	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	F3	left plantar foot	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFO92CG	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F32Fotl	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
F33Fotl.140406	GTCAACGCGATG	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	left_sole_of_foot	Human Skin	source	NA	n	11/14/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	left sole of foot	FKB0RMH	33	mimarks-survey	NA	FMA:Left foot surface	NA	NA	0	UBERON:zone of skin of foot	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F3	sample231	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F33Fotl	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	F3	left plantar foot	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F33Fotl	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
F34Fotl.140856	GTCTCTCTACGC	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	left_sole_of_foot	Human Skin	source	NA	n	11/14/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	left sole of foot	FKB0RMH	33	mimarks-survey	NA	FMA:Left foot surface	NA	NA	0	UBERON:zone of skin of foot	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F3	sample232	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F34Fotl	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	F3	left plantar foot	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F34Fotl	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M11Fotl.140711	ACCAGACGATGC	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	left_sole_of_foot	Human Skin	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	left sole of foot	FFLHOYS	36	mimarks-survey	NA	FMA:Left foot surface	NA	NA	0	UBERON:zone of skin of foot	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M1	sample233	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M11Fotl	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	36	M1	left plantar foot	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M11Fotl	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M12Fotl.140774	ACTCTTCTAGAG	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	left_sole_of_foot	Human Skin	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	left sole of foot	FFLHOYS	36	mimarks-survey	NA	FMA:Left foot surface	NA	NA	0	UBERON:zone of skin of foot	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M1	sample234	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M12Fotl	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	36	M1	left plantar foot	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M12Fotl	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M13Fotl.140685	GAAGAGTGATCA	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	left_sole_of_foot	Human Skin	source	NA	n	11/14/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	left sole of foot	FKB0RMH	36	mimarks-survey	NA	FMA:Left foot surface	NA	NA	0	UBERON:zone of skin of foot	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M1	sample235	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M13Fotl	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	36	M1	left plantar foot	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M13Fotl	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M14Fotl.140639	GACGATATCGCG	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	left_sole_of_foot	Human Skin	source	NA	n	11/14/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	left sole of foot	FKB0RMH	36	mimarks-survey	NA	FMA:Left foot surface	NA	NA	0	UBERON:zone of skin of foot	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M1	sample236	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M14Fotl	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	36	M1	left plantar foot	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M14Fotl	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M21Fotl.140499	CATGGCTACACA	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	left_sole_of_foot	Human Skin	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	left sole of foot	FFO92CG	36	mimarks-survey	NA	FMA:Left foot surface	NA	NA	0	UBERON:zone of skin of foot	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M2	sample237	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M21Fotl	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	36	M2	left plantar foot	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFO92CG	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M21Fotl	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M22Fotl.140555	CGCACTCTAGAA	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	left_sole_of_foot	Human Skin	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	left sole of foot	FFO92CG	36	mimarks-survey	NA	FMA:Left foot surface	NA	NA	0	UBERON:zone of skin of foot	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M2	sample238	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M22Fotl	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	36	M2	left plantar foot	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFO92CG	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M22Fotl	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M23Fotl.140684	GATCTTCAGTAC	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	left_sole_of_foot	Human Skin	source	NA	n	11/14/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	left sole of foot	FKB0RMH	36	mimarks-survey	NA	FMA:Left foot surface	NA	NA	0	UBERON:zone of skin of foot	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M2	sample239	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M23Fotl	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	36	M2	left plantar foot	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M23Fotl	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M24Fotl.140320	GCAGCCGAGTAT	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	left_sole_of_foot	Human Skin	source	NA	n	11/14/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	left sole of foot	FKB0RMH	36	mimarks-survey	NA	FMA:Left foot surface	NA	NA	0	UBERON:zone of skin of foot	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M2	sample240	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M24Fotl	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	36	M2	left plantar foot	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M24Fotl	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M31Fotl.140712	CTAACGCAGTCA	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	left_sole_of_foot	Human Skin	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	left sole of foot	FFLHOYS	33	mimarks-survey	NA	FMA:Left foot surface	NA	NA	0	UBERON:zone of skin of foot	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M3	sample241	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M31Fotl	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	M3	left plantar foot	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M31Fotl	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M32Fotl.140843	CTGACACGACAG	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	left_sole_of_foot	Human Skin	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	left sole of foot	FFLHOYS	33	mimarks-survey	NA	FMA:Left foot surface	NA	NA	0	UBERON:zone of skin of foot	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M3	sample242	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M32Fotl	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	M3	left plantar foot	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M32Fotl	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M33Fotl.140701	GGATCGCAGATC	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	left_sole_of_foot	Human Skin	source	NA	n	11/14/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	left sole of foot	FKB0RMH	33	mimarks-survey	NA	FMA:Left foot surface	NA	NA	0	UBERON:zone of skin of foot	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M3	sample243	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M33Fotl	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	M3	left plantar foot	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M33Fotl	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M34Fotl.140468	GTAGACTGCGTG	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	left_sole_of_foot	Human Skin	source	NA	n	11/14/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	left sole of foot	FKB0RMH	33	mimarks-survey	NA	FMA:Left foot surface	NA	NA	0	UBERON:zone of skin of foot	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M3	sample244	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M34Fotl	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	M3	left plantar foot	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M34Fotl	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M41Fotl.140801	GACCGAGCTATG	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	left_sole_of_foot	Human Skin	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	left sole of foot	FFO92CG	31	mimarks-survey	NA	FMA:Left foot surface	NA	NA	0	UBERON:zone of skin of foot	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M4	sample245	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M41Fotl	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	31	M4	left plantar foot	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFO92CG	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M41Fotl	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M42Fotl.140742	GATACGTCCTGA	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	left_sole_of_foot	Human Skin	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	left sole of foot	FFO92CG	31	mimarks-survey	NA	FMA:Left foot surface	NA	NA	0	UBERON:zone of skin of foot	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M4	sample246	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M42Fotl	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	31	M4	left plantar foot	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFO92CG	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M42Fotl	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M43Fotl.140659	GTGTCTACATTG	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	left_sole_of_foot	Human Skin	source	NA	n	11/14/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	left sole of foot	FKB0RMH	31	mimarks-survey	NA	FMA:Left foot surface	NA	NA	0	UBERON:zone of skin of foot	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M4	sample247	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M43Fotl	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	31	M4	left plantar foot	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M43Fotl	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M44Fotl.140595	TACACGATCTAC	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	left_sole_of_foot	Human Skin	source	NA	n	11/14/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	left sole of foot	FKB0RMH	31	mimarks-survey	NA	FMA:Left foot surface	NA	NA	0	UBERON:zone of skin of foot	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M4	sample248	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M44Fotl	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	31	M4	left plantar foot	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M44Fotl	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M53Fotl.140376	TACTGGACGCGA	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	left_sole_of_foot	Human Skin	source	NA	n	11/14/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	left sole of foot	FKB0RMH	60	mimarks-survey	NA	FMA:Left foot surface	NA	NA	0	UBERON:zone of skin of foot	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M5	sample249	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M53Fotl	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	60	M5	left plantar foot	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M53Fotl	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M54Fotl.140823	TAGCTCGTAACT	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	left_sole_of_foot	Human Skin	source	NA	n	11/14/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	left sole of foot	FKB0RMH	60	mimarks-survey	NA	FMA:Left foot surface	NA	NA	0	UBERON:zone of skin of foot	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M5	sample250	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M54Fotl	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	60	M5	left plantar foot	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M54Fotl	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M63Fotl.140731	TATGCGAGGTCG	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	left_sole_of_foot	Human Skin	source	NA	n	11/14/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	left sole of foot	FKB0RMH	NA	mimarks-survey	NA	FMA:Left foot surface	NA	NA	0	UBERON:zone of skin of foot	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M6	sample251	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M63Fotl	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	NA	M6	left plantar foot	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M63Fotl	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M64Fotl.140777	TCAGCCATGACA	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	left_sole_of_foot	Human Skin	source	NA	n	11/14/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	left sole of foot	FKB0RMH	NA	mimarks-survey	NA	FMA:Left foot surface	NA	NA	0	UBERON:zone of skin of foot	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M6	sample252	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M64Fotl	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	NA	M6	left plantar foot	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M64Fotl	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
F11Mout.140583	AGCACACCTACA	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	mouth	Human Mouth	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	mouth	FFLHOYS	33	mimarks-survey	NA	FMA:Oral cavity	NA	NA	0	UBERON:mouth	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F1	sample253	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:oral cavity	UBERON:saliva	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F11Mout	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	F1	oral cavity	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human oral metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	447426	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F11Mout	ENVO:saliva	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human oral metagenome
F12Mout.140450	AGTGGATGCTCT	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	mouth	Human Mouth	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	mouth	FFLHOYS	33	mimarks-survey	NA	FMA:Oral cavity	NA	NA	0	UBERON:mouth	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F1	sample254	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:oral cavity	UBERON:saliva	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F12Mout	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	F1	oral cavity	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human oral metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	447426	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F12Mout	ENVO:saliva	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human oral metagenome
F21Mout.140380	ATCTTAGACTGC	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	mouth	Human Mouth	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	mouth	FFLHOYS	31	mimarks-survey	NA	FMA:Oral cavity	NA	NA	0	UBERON:mouth	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F2	sample255	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:oral cavity	UBERON:saliva	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F21Mout	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	31	F2	oral cavity	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human oral metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	447426	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F21Mout	ENVO:saliva	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human oral metagenome
F22Mout.140788	CACATTGTGAGC	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	mouth	Human Mouth	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	mouth	FFLHOYS	31	mimarks-survey	NA	FMA:Oral cavity	NA	NA	0	UBERON:mouth	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F2	sample256	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:oral cavity	UBERON:saliva	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F22Mout	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	31	F2	oral cavity	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human oral metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	447426	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F22Mout	ENVO:saliva	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human oral metagenome
F31Mout.140605	GCACTCGTTAGA	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	mouth	Human Mouth	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	mouth	FFO92CG	33	mimarks-survey	NA	FMA:Oral cavity	NA	NA	0	UBERON:mouth	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F3	sample257	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:oral cavity	UBERON:saliva	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F31Mout	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	F3	oral cavity	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human oral metagenome	NA	ENVO:human-associated habitat	FLX	FFO92CG	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	447426	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F31Mout	ENVO:saliva	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human oral metagenome
F32Mout.140603	GCTCAGTGCAGA	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	mouth	Human Mouth	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	mouth	FFO92CG	33	mimarks-survey	NA	FMA:Oral cavity	NA	NA	0	UBERON:mouth	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F3	sample258	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:oral cavity	UBERON:saliva	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F32Mout	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	F3	oral cavity	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human oral metagenome	NA	ENVO:human-associated habitat	FLX	FFO92CG	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	447426	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F32Mout	ENVO:saliva	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human oral metagenome
M11Mout.140852	ACAGCAGTGGTC	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	mouth	Human Mouth	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	mouth	FFLHOYS	36	mimarks-survey	NA	FMA:Oral cavity	NA	NA	0	UBERON:mouth	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M1	sample259	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:oral cavity	UBERON:saliva	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M11Mout	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	36	M1	oral cavity	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human oral metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	447426	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M11Mout	ENVO:saliva	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human oral metagenome
M12Mout.140350	ACGTTAGCACAC	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	mouth	Human Mouth	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	mouth	FFLHOYS	36	mimarks-survey	NA	FMA:Oral cavity	NA	NA	0	UBERON:mouth	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M1	sample260	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:oral cavity	UBERON:saliva	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M12Mout	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	36	M1	oral cavity	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human oral metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	447426	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M12Mout	ENVO:saliva	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human oral metagenome
M21Mout.140501	CATAGCGAGTTC	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	mouth	Human Mouth	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	mouth	FFO92CG	36	mimarks-survey	NA	FMA:Oral cavity	NA	NA	0	UBERON:mouth	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M2	sample261	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:oral cavity	UBERON:saliva	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M21Mout	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	36	M2	oral cavity	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human oral metagenome	NA	ENVO:human-associated habitat	FLX	FFO92CG	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	447426	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M21Mout	ENVO:saliva	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human oral metagenome
M22Mout.140541	CGAGGCTCAGTA	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	mouth	Human Mouth	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	mouth	FFO92CG	36	mimarks-survey	NA	FMA:Oral cavity	NA	NA	0	UBERON:mouth	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M2	sample262	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:oral cavity	UBERON:saliva	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M22Mout	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	36	M2	oral cavity	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human oral metagenome	NA	ENVO:human-associated habitat	FLX	FFO92CG	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	447426	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M22Mout	ENVO:saliva	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human oral metagenome
M31Mout.140342	CGTCGATCTCTC	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	mouth	Human Mouth	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	mouth	FFLHOYS	33	mimarks-survey	NA	FMA:Oral cavity	NA	NA	0	UBERON:mouth	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M3	sample263	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:oral cavity	UBERON:saliva	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M31Mout	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	M3	oral cavity	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human oral metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	447426	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M31Mout	ENVO:saliva	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human oral metagenome
M32Mout.140657	CTCCTACTGTCT	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	mouth	Human Mouth	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	mouth	FFLHOYS	33	mimarks-survey	NA	FMA:Oral cavity	NA	NA	0	UBERON:mouth	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M3	sample264	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:oral cavity	UBERON:saliva	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M32Mout	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	M3	oral cavity	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human oral metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	447426	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M32Mout	ENVO:saliva	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human oral metagenome
M41Mout.140367	GAAGCTACTGTC	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	mouth	Human Mouth	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	mouth	FFO92CG	31	mimarks-survey	NA	FMA:Oral cavity	NA	NA	0	UBERON:mouth	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M4	sample265	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:oral cavity	UBERON:saliva	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M41Mout	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	31	M4	oral cavity	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human oral metagenome	NA	ENVO:human-associated habitat	FLX	FFO92CG	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	447426	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M41Mout	ENVO:saliva	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human oral metagenome
M42Mout.140653	GAGCAGATGCCT	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	mouth	Human Mouth	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	mouth	FFO92CG	31	mimarks-survey	NA	FMA:Oral cavity	NA	NA	0	UBERON:mouth	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M4	sample266	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:oral cavity	UBERON:saliva	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M42Mout	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	31	M4	oral cavity	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human oral metagenome	NA	ENVO:human-associated habitat	FLX	FFO92CG	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	447426	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M42Mout	ENVO:saliva	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human oral metagenome
F11Navl.140566	AGCACGAGCCTA	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	navel	Human Skin	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	navel	FFLHOYS	33	mimarks-survey	NA	FMA:Umbilicus	NA	NA	0	UBERON:zone of skin of abdomen	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F1	sample267	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F11Navl	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	F1	umbilicus	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F11Navl	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
F12Navl.140828	AGTGTCACGGTG	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	navel	Human Skin	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	navel	FFLHOYS	33	mimarks-survey	NA	FMA:Umbilicus	NA	NA	0	UBERON:zone of skin of abdomen	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F1	sample268	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F12Navl	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	F1	umbilicus	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F12Navl	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
F21Navl.140840	ATGACCATCGTG	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	navel	Human Skin	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	navel	FFLHOYS	31	mimarks-survey	NA	FMA:Umbilicus	NA	NA	0	UBERON:zone of skin of abdomen	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F2	sample269	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F21Navl	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	31	F2	umbilicus	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F21Navl	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
F22Navl.140838	CACGACAGGCTA	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	navel	Human Skin	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	navel	FFLHOYS	31	mimarks-survey	NA	FMA:Umbilicus	NA	NA	0	UBERON:zone of skin of abdomen	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F2	sample270	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F22Navl	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	31	F2	umbilicus	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F22Navl	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
F32Navl.140293	GCTCGCTACTTC	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	navel	Human Skin	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	navel	FFO92CG	33	mimarks-survey	NA	FMA:Umbilicus	NA	NA	0	UBERON:zone of skin of abdomen	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F3	sample271	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F32Navl	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	F3	umbilicus	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFO92CG	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F32Navl	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M11Navl.140483	ACAGCTAGCTTG	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	navel	Human Skin	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	navel	FFLHOYS	36	mimarks-survey	NA	FMA:Umbilicus	NA	NA	0	UBERON:zone of skin of abdomen	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M1	sample272	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M11Navl	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	36	M1	umbilicus	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M11Navl	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M12Navl.140334	ACTACAGCCTAT	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	navel	Human Skin	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	navel	FFLHOYS	36	mimarks-survey	NA	FMA:Umbilicus	NA	NA	0	UBERON:zone of skin of abdomen	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M1	sample273	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M12Navl	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	36	M1	umbilicus	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M12Navl	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M21Navl.140850	CATATACTCGCA	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	navel	Human Skin	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	navel	FFO92CG	36	mimarks-survey	NA	FMA:Umbilicus	NA	NA	0	UBERON:zone of skin of abdomen	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M2	sample274	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M21Navl	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	36	M2	umbilicus	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFO92CG	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M21Navl	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M22Navl.140637	CGAGTCTAGTTG	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	navel	Human Skin	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	navel	FFO92CG	36	mimarks-survey	NA	FMA:Umbilicus	NA	NA	0	UBERON:zone of skin of abdomen	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M2	sample275	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M22Navl	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	36	M2	umbilicus	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFO92CG	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M22Navl	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M31Navl.140871	CGTGACAATGTC	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	navel	Human Skin	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	navel	FFLHOYS	33	mimarks-survey	NA	FMA:Umbilicus	NA	NA	0	UBERON:zone of skin of abdomen	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M3	sample276	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M31Navl	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	M3	umbilicus	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M31Navl	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M41Navl.140280	GAAGTCTCGCAT	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	navel	Human Skin	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	navel	FFO92CG	31	mimarks-survey	NA	FMA:Umbilicus	NA	NA	0	UBERON:zone of skin of abdomen	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M4	sample277	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M41Navl	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	31	M4	umbilicus	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFO92CG	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M41Navl	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M42Navl.140564	GAGCATTCTCTA	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	navel	Human Skin	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	navel	FFO92CG	31	mimarks-survey	NA	FMA:Umbilicus	NA	NA	0	UBERON:zone of skin of abdomen	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M4	sample278	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M42Navl	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	31	M4	umbilicus	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFO92CG	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M42Navl	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
F13Nost.140811	GACTGCATCTTA	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	nostrils	Human Skin	source	NA	n	11/14/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	nostrils	FKB0RMH	33	mimarks-survey	NA	FMA:Nares	NA	NA	0	UBERON:nostrils	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F1	sample279	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:nostril	UBERON:mucus	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F13Nost	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	F1	nares	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human nose metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	646099	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F13Nost	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human nose metagenome
F14Nost.140522	GAGTAGCTCGTG	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	nostrils	Human Skin	source	NA	n	11/14/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	nostrils	FKB0RMH	33	mimarks-survey	NA	FMA:Nares	NA	NA	0	UBERON:nostrils	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F1	sample280	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:nostril	UBERON:mucus	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F14Nost	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	F1	nares	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human nose metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	646099	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F14Nost	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human nose metagenome
F23Nost.140510	GCATTGCGTGAG	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	nostrils	Human Skin	source	NA	n	11/14/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	nostrils	FKB0RMH	31	mimarks-survey	NA	FMA:Nares	NA	NA	0	UBERON:nostrils	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F2	sample281	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:nostril	UBERON:mucus	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F23Nost	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	31	F2	nares	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human nose metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	646099	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F23Nost	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human nose metagenome
F24Nost.140792	GCTAGATGCCAG	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	nostrils	Human Skin	source	NA	n	11/14/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	nostrils	FKB0RMH	31	mimarks-survey	NA	FMA:Nares	NA	NA	0	UBERON:nostrils	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F2	sample282	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:nostril	UBERON:mucus	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F24Nost	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	31	F2	nares	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human nose metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	646099	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F24Nost	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human nose metagenome
F33Nost.140487	GTATGACTGGCT	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	nostrils	Human Skin	source	NA	n	11/14/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	nostrils	FKB0RMH	33	mimarks-survey	NA	FMA:Nares	NA	NA	0	UBERON:nostrils	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F3	sample283	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:nostril	UBERON:mucus	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F33Nost	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	F3	nares	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human nose metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	646099	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F33Nost	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human nose metagenome
F34Nost.140282	GTCTACACACAT	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	nostrils	Human Skin	source	NA	n	11/14/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	nostrils	FKB0RMH	33	mimarks-survey	NA	FMA:Nares	NA	NA	0	UBERON:nostrils	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F3	sample284	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:nostril	UBERON:mucus	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F34Nost	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	F3	nares	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human nose metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	646099	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F34Nost	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human nose metagenome
M13Nost.140290	CTTGTGTCGATA	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	nostrils	Human Skin	source	NA	n	11/14/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	nostrils	FKB0RMH	36	mimarks-survey	NA	FMA:Nares	NA	NA	0	UBERON:nostrils	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M1	sample285	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:nostril	UBERON:mucus	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M13Nost	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	36	M1	nares	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human nose metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	646099	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M13Nost	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human nose metagenome
M14Nost.140445	GACCACTACGAT	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	nostrils	Human Skin	source	NA	n	11/14/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	nostrils	FKB0RMH	36	mimarks-survey	NA	FMA:Nares	NA	NA	0	UBERON:nostrils	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M1	sample286	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:nostril	UBERON:mucus	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M14Nost	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	36	M1	nares	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human nose metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	646099	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M14Nost	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human nose metagenome
M23Nost.140669	GATCGTCCAGAT	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	nostrils	Human Skin	source	NA	n	11/14/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	nostrils	FKB0RMH	36	mimarks-survey	NA	FMA:Nares	NA	NA	0	UBERON:nostrils	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M2	sample287	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:nostril	UBERON:mucus	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M23Nost	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	36	M2	nares	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human nose metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	646099	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M23Nost	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human nose metagenome
M24Nost.140862	GCACTCGTTAGA	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	nostrils	Human Skin	source	NA	n	11/14/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	nostrils	FKB0RMH	36	mimarks-survey	NA	FMA:Nares	NA	NA	0	UBERON:nostrils	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M2	sample288	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:nostril	UBERON:mucus	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M24Nost	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	36	M2	nares	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human nose metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	646099	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M24Nost	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human nose metagenome
M33Nost.140351	GCTTCATAGTGT	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	nostrils	Human Skin	source	NA	n	11/14/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	nostrils	FKB0RMH	33	mimarks-survey	NA	FMA:Nares	NA	NA	0	UBERON:nostrils	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M3	sample289	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:nostril	UBERON:mucus	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M33Nost	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	M3	nares	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human nose metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	646099	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M33Nost	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human nose metagenome
M34Nost.140408	GTACAAGAGTGA	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	nostrils	Human Skin	source	NA	n	11/14/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	nostrils	FKB0RMH	33	mimarks-survey	NA	FMA:Nares	NA	NA	0	UBERON:nostrils	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M3	sample290	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:nostril	UBERON:mucus	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M34Nost	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	M3	nares	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human nose metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	646099	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M34Nost	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human nose metagenome
M43Nost.140345	GTGCACATTATC	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	nostrils	Human Skin	source	NA	n	11/14/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	nostrils	FKB0RMH	31	mimarks-survey	NA	FMA:Nares	NA	NA	0	UBERON:nostrils	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M4	sample291	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:nostril	UBERON:mucus	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M43Nost	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	31	M4	nares	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human nose metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	646099	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M43Nost	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human nose metagenome
M44Nost.140343	TAACTCTGATGC	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	nostrils	Human Skin	source	NA	n	11/14/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	nostrils	FKB0RMH	31	mimarks-survey	NA	FMA:Nares	NA	NA	0	UBERON:nostrils	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M4	sample292	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:nostril	UBERON:mucus	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M44Nost	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	31	M4	nares	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human nose metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	646099	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M44Nost	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human nose metagenome
M53Nost.140611	TACTAATCTGCG	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	nostrils	Human Skin	source	NA	n	11/14/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	nostrils	FKB0RMH	60	mimarks-survey	NA	FMA:Nares	NA	NA	0	UBERON:nostrils	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M5	sample293	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:nostril	UBERON:mucus	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M53Nost	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	60	M5	nares	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human nose metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	646099	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M53Nost	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human nose metagenome
M54Nost.140835	TAGCCTCTCTGC	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	nostrils	Human Skin	source	NA	n	11/14/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	nostrils	FKB0RMH	60	mimarks-survey	NA	FMA:Nares	NA	NA	0	UBERON:nostrils	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M5	sample294	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:nostril	UBERON:mucus	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M54Nost	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	60	M5	nares	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human nose metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	646099	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M54Nost	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human nose metagenome
M63Nost.140870	TATCGCGCGATA	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	nostrils	Human Skin	source	NA	n	11/14/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	nostrils	FKB0RMH	NA	mimarks-survey	NA	FMA:Nares	NA	NA	0	UBERON:nostrils	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M6	sample295	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:nostril	UBERON:mucus	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M63Nost	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	NA	M6	nares	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human nose metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	646099	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M63Nost	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human nose metagenome
M64Nost.140402	TCACTTCTCGCT	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	nostrils	Human Skin	source	NA	n	11/14/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	nostrils	FKB0RMH	NA	mimarks-survey	NA	FMA:Nares	NA	NA	0	UBERON:nostrils	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M6	sample296	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:nostril	UBERON:mucus	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M64Nost	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	NA	M6	nares	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human nose metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	646099	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M64Nost	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human nose metagenome
F13Ewax.140760	GAGACAGCTTGC	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	outer_ear_canal_earwax	Human Skin	source	NA	n	11/14/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	outer ear canal/earwax	FKB0RMH	33	mimarks-survey	NA	FMA:External auditory canal	NA	NA	0	UBERON:zone of skin of outer ear	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F1	sample297	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:external auditory canal	UBERON:cerumen	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F13Ewax	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	F1	external auditory canal	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	646099	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F13Ewax	ENVO:ear wax	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human metagenome
F14Ewax.140371	GAGTGGTAGAGA	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	outer_ear_canal_earwax	Human Skin	source	NA	n	11/14/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	outer ear canal/earwax	FKB0RMH	33	mimarks-survey	NA	FMA:External auditory canal	NA	NA	0	UBERON:zone of skin of outer ear	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F1	sample298	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:external auditory canal	UBERON:cerumen	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F14Ewax	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	F1	external auditory canal	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	646099	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F14Ewax	ENVO:ear wax	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human metagenome
F23Ewax.140304	GCGACTTGTGTA	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	outer_ear_canal_earwax	Human Skin	source	NA	n	11/14/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	outer ear canal/earwax	FKB0RMH	31	mimarks-survey	NA	FMA:External auditory canal	NA	NA	0	UBERON:zone of skin of outer ear	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F2	sample299	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:external auditory canal	UBERON:cerumen	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F23Ewax	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	31	F2	external auditory canal	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	646099	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F23Ewax	ENVO:ear wax	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human metagenome
F24Ewax.140807	GCTCAGTGCAGA	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	outer_ear_canal_earwax	Human Skin	source	NA	n	11/14/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	outer ear canal/earwax	FKB0RMH	31	mimarks-survey	NA	FMA:External auditory canal	NA	NA	0	UBERON:zone of skin of outer ear	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F2	sample300	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:external auditory canal	UBERON:cerumen	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F24Ewax	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	31	F2	external auditory canal	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	646099	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F24Ewax	ENVO:ear wax	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human metagenome
F33Ewax.140872	GTCACGACTATT	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	outer_ear_canal_earwax	Human Skin	source	NA	n	11/14/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	outer ear canal/earwax	FKB0RMH	33	mimarks-survey	NA	FMA:External auditory canal	NA	NA	0	UBERON:zone of skin of outer ear	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F3	sample301	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:external auditory canal	UBERON:cerumen	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F33Ewax	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	F3	external auditory canal	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	646099	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F33Ewax	ENVO:ear wax	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human metagenome
F34Ewax.140285	GTCTGACAGTTG	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	outer_ear_canal_earwax	Human Skin	source	NA	n	11/14/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	outer ear canal/earwax	FKB0RMH	33	mimarks-survey	NA	FMA:External auditory canal	NA	NA	0	UBERON:zone of skin of outer ear	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F3	sample302	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:external auditory canal	UBERON:cerumen	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F34Ewax	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	F3	external auditory canal	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	646099	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F34Ewax	ENVO:ear wax	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human metagenome
M13Ewax.140391	GAAGCTACTGTC	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	outer_ear_canal_earwax	Human Skin	source	NA	n	11/14/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	outer ear canal/earwax	FKB0RMH	36	mimarks-survey	NA	FMA:External auditory canal	NA	NA	0	UBERON:zone of skin of outer ear	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M1	sample303	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:external auditory canal	UBERON:cerumen	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M13Ewax	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	36	M1	external auditory canal	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	646099	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M13Ewax	ENVO:ear wax	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human metagenome
M14Ewax.140489	GACGCAGTAGCT	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	outer_ear_canal_earwax	Human Skin	source	NA	n	11/14/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	outer ear canal/earwax	FKB0RMH	36	mimarks-survey	NA	FMA:External auditory canal	NA	NA	0	UBERON:zone of skin of outer ear	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M1	sample304	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:external auditory canal	UBERON:cerumen	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M14Ewax	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	36	M1	external auditory canal	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	646099	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M14Ewax	ENVO:ear wax	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human metagenome
M23Ewax.140338	GATGATCGCCGA	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	outer_ear_canal_earwax	Human Skin	source	NA	n	11/14/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	outer ear canal/earwax	FKB0RMH	36	mimarks-survey	NA	FMA:External auditory canal	NA	NA	0	UBERON:zone of skin of outer ear	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M2	sample305	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:external auditory canal	UBERON:cerumen	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M23Ewax	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	36	M2	external auditory canal	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	646099	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M23Ewax	ENVO:ear wax	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human metagenome
M24Ewax.140460	GCAGGATAGATA	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	outer_ear_canal_earwax	Human Skin	source	NA	n	11/14/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	outer ear canal/earwax	FKB0RMH	36	mimarks-survey	NA	FMA:External auditory canal	NA	NA	0	UBERON:zone of skin of outer ear	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M2	sample306	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:external auditory canal	UBERON:cerumen	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M24Ewax	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	36	M2	external auditory canal	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	646099	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M24Ewax	ENVO:ear wax	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human metagenome
M33Ewax.140613	GGCAGTGTATCG	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	outer_ear_canal_earwax	Human Skin	source	NA	n	11/14/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	outer ear canal/earwax	FKB0RMH	33	mimarks-survey	NA	FMA:External auditory canal	NA	NA	0	UBERON:zone of skin of outer ear	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M3	sample307	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:external auditory canal	UBERON:cerumen	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M33Ewax	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	M3	external auditory canal	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	646099	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M33Ewax	ENVO:ear wax	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human metagenome
M34Ewax.140864	GTAGAGCTGTTC	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	outer_ear_canal_earwax	Human Skin	source	NA	n	11/14/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	outer ear canal/earwax	FKB0RMH	33	mimarks-survey	NA	FMA:External auditory canal	NA	NA	0	UBERON:zone of skin of outer ear	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M3	sample308	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:external auditory canal	UBERON:cerumen	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M34Ewax	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	M3	external auditory canal	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	646099	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M34Ewax	ENVO:ear wax	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human metagenome
M43Ewax.140650	GTGTGCTATCAG	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	outer_ear_canal_earwax	Human Skin	source	NA	n	11/14/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	outer ear canal/earwax	FKB0RMH	31	mimarks-survey	NA	FMA:External auditory canal	NA	NA	0	UBERON:zone of skin of outer ear	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M4	sample309	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:external auditory canal	UBERON:cerumen	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M43Ewax	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	31	M4	external auditory canal	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	646099	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M43Ewax	ENVO:ear wax	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human metagenome
M44Ewax.140411	TACAGATGGCTC	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	outer_ear_canal_earwax	Human Skin	source	NA	n	11/14/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	outer ear canal/earwax	FKB0RMH	31	mimarks-survey	NA	FMA:External auditory canal	NA	NA	0	UBERON:zone of skin of outer ear	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M4	sample310	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:external auditory canal	UBERON:cerumen	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M44Ewax	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	31	M4	external auditory canal	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	646099	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M44Ewax	ENVO:ear wax	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human metagenome
M53Ewax.140448	TACTTACTGCAG	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	outer_ear_canal_earwax	Human Skin	source	NA	n	11/14/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	outer ear canal/earwax	FKB0RMH	60	mimarks-survey	NA	FMA:External auditory canal	NA	NA	0	UBERON:zone of skin of outer ear	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M5	sample311	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:external auditory canal	UBERON:cerumen	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M53Ewax	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	60	M5	external auditory canal	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	646099	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M53Ewax	ENVO:ear wax	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human metagenome
M54Ewax.140661	TAGCTGAGTCCA	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	outer_ear_canal_earwax	Human Skin	source	NA	n	11/14/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	outer ear canal/earwax	FKB0RMH	60	mimarks-survey	NA	FMA:External auditory canal	NA	NA	0	UBERON:zone of skin of outer ear	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M5	sample312	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:external auditory canal	UBERON:cerumen	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M54Ewax	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	60	M5	external auditory canal	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	646099	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M54Ewax	ENVO:ear wax	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human metagenome
M63Ewax.140384	TATGGCACACAC	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	outer_ear_canal_earwax	Human Skin	source	NA	n	11/14/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	outer ear canal/earwax	FKB0RMH	NA	mimarks-survey	NA	FMA:External auditory canal	NA	NA	0	UBERON:zone of skin of outer ear	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M6	sample313	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:external auditory canal	UBERON:cerumen	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M63Ewax	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	NA	M6	external auditory canal	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	646099	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M63Ewax	ENVO:ear wax	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human metagenome
M64Ewax.140550	TCAGCTCAACTA	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	outer_ear_canal_earwax	Human Skin	source	NA	n	11/14/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	outer ear canal/earwax	FKB0RMH	NA	mimarks-survey	NA	FMA:External auditory canal	NA	NA	0	UBERON:zone of skin of outer ear	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M6	sample314	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:external auditory canal	UBERON:cerumen	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M64Ewax	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	NA	M6	external auditory canal	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	646099	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M64Ewax	ENVO:ear wax	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human metagenome
F12Aptr.140530	AGCTTGACAGCT	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	right_armpit	Human Skin	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	right armpit	FFLHOYS	33	mimarks-survey	NA	FMA:Right axilla	NA	NA	0	UBERON:skin of arm	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F1	sample316	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F12Aptr	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	F1	right axilla	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F12Aptr	ENVO:sweat	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
F21Aptr.140500	ATATGCCAGTGC	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	right_armpit	Human Skin	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	right armpit	FFLHOYS	31	mimarks-survey	NA	FMA:Right axilla	NA	NA	0	UBERON:skin of arm	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F2	sample317	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F21Aptr	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	31	F2	right axilla	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F21Aptr	ENVO:sweat	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
F22Aptr.140609	ATGTCACCGTGA	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	right_armpit	Human Skin	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	right armpit	FFLHOYS	31	mimarks-survey	NA	FMA:Right axilla	NA	NA	0	UBERON:skin of arm	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F2	sample318	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F22Aptr	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	31	F2	right axilla	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F22Aptr	ENVO:sweat	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
F31Aptr.140458	GATCCGACACTA	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	right_armpit	Human Skin	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	right armpit	FFO92CG	33	mimarks-survey	NA	FMA:Right axilla	NA	NA	0	UBERON:skin of arm	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F3	sample319	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F31Aptr	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	F3	right axilla	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFO92CG	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F31Aptr	ENVO:sweat	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
F32Aptr.140339	GCCAGAGTCGTA	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	right_armpit	Human Skin	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	right armpit	FFO92CG	33	mimarks-survey	NA	FMA:Right axilla	NA	NA	0	UBERON:skin of arm	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F3	sample320	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F32Aptr	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	F3	right axilla	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFO92CG	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F32Aptr	ENVO:sweat	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M11Aptr.140378	AACTCGTCGATG	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	right_armpit	Human Skin	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	right armpit	FFLHOYS	36	mimarks-survey	NA	FMA:Right axilla	NA	NA	0	UBERON:skin of arm	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M1	sample321	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M11Aptr	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	36	M1	right axilla	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M11Aptr	ENVO:sweat	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M12Aptr.140800	ACGACGTCTTAG	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	right_armpit	Human Skin	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	right armpit	FFLHOYS	36	mimarks-survey	NA	FMA:Right axilla	NA	NA	0	UBERON:skin of arm	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M1	sample322	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M12Aptr	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	36	M1	right axilla	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M12Aptr	ENVO:sweat	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M21Aptr.140329	CAGATCGGATCG	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	right_armpit	Human Skin	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	right armpit	FFO92CG	36	mimarks-survey	NA	FMA:Right axilla	NA	NA	0	UBERON:skin of arm	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M2	sample323	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M21Aptr	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	36	M2	right axilla	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFO92CG	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M21Aptr	ENVO:sweat	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M22Aptr.140827	CCAGATGATCGT	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	right_armpit	Human Skin	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	right armpit	FFO92CG	36	mimarks-survey	NA	FMA:Right axilla	NA	NA	0	UBERON:skin of arm	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M2	sample324	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M22Aptr	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	36	M2	right axilla	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFO92CG	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M22Aptr	ENVO:sweat	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M31Aptr.140474	CGCGTAACTGTA	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	right_armpit	Human Skin	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	right armpit	FFLHOYS	33	mimarks-survey	NA	FMA:Right axilla	NA	NA	0	UBERON:skin of arm	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M3	sample325	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M31Aptr	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	M3	right axilla	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M31Aptr	ENVO:sweat	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M32Aptr.140847	CTACTGATATCG	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	right_armpit	Human Skin	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	right armpit	FFLHOYS	33	mimarks-survey	NA	FMA:Right axilla	NA	NA	0	UBERON:skin of arm	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M3	sample326	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M32Aptr	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	M3	right axilla	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M32Aptr	ENVO:sweat	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M41Aptr.140772	CTGGACTCATAG	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	right_armpit	Human Skin	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	right armpit	FFO92CG	31	mimarks-survey	NA	FMA:Right axilla	NA	NA	0	UBERON:skin of arm	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M4	sample327	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M41Aptr	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	31	M4	right axilla	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFO92CG	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M41Aptr	ENVO:sweat	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M42Aptr.140377	GACGTTGCACAG	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	right_armpit	Human Skin	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	right armpit	FFO92CG	31	mimarks-survey	NA	FMA:Right axilla	NA	NA	0	UBERON:skin of arm	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M4	sample328	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M42Aptr	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	31	M4	right axilla	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFO92CG	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M42Aptr	ENVO:sweat	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
F11Kner.140478	AGAACACGTCTC	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	right_back_of_knees	Human Skin	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	right back of knees	FFLHOYS	33	mimarks-survey	NA	FMA:Right popliteal fossa	NA	NA	0	UBERON:zone of skin of knee	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F1	sample329	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F11Kner	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	F1	right popliteal fossa	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F11Kner	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
F12Kner.140545	AGGCTACACGAC	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	right_back_of_knees	Human Skin	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	right back of knees	FFLHOYS	33	mimarks-survey	NA	FMA:Right popliteal fossa	NA	NA	0	UBERON:zone of skin of knee	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F1	sample330	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F12Kner	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	F1	right popliteal fossa	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F12Kner	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
F21Kner.140678	ATCACTAGTCAC	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	right_back_of_knees	Human Skin	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	right back of knees	FFLHOYS	31	mimarks-survey	NA	FMA:Right popliteal fossa	NA	NA	0	UBERON:zone of skin of knee	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F2	sample331	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F21Kner	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	31	F2	right popliteal fossa	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F21Kner	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
F22Kner.140705	ATGTGTCGACTT	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	right_back_of_knees	Human Skin	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	right back of knees	FFLHOYS	31	mimarks-survey	NA	FMA:Right popliteal fossa	NA	NA	0	UBERON:zone of skin of knee	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F2	sample332	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F22Kner	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	31	F2	right popliteal fossa	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F22Kner	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
F31Kner.140761	GATCGTCCAGAT	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	right_back_of_knees	Human Skin	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	right back of knees	FFO92CG	33	mimarks-survey	NA	FMA:Right popliteal fossa	NA	NA	0	UBERON:zone of skin of knee	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F3	sample333	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F31Kner	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	F3	right popliteal fossa	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFO92CG	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F31Kner	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
F32Kner.140397	GCGACTTGTGTA	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	right_back_of_knees	Human Skin	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	right back of knees	FFO92CG	33	mimarks-survey	NA	FMA:Right popliteal fossa	NA	NA	0	UBERON:zone of skin of knee	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F3	sample334	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F32Kner	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	F3	right popliteal fossa	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFO92CG	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F32Kner	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M11Kner.140517	AAGAGATGTCGA	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	right_back_of_knees	Human Skin	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	right back of knees	FFLHOYS	36	mimarks-survey	NA	FMA:Right popliteal fossa	NA	NA	0	UBERON:zone of skin of knee	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M1	sample335	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M11Kner	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	36	M1	right popliteal fossa	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M11Kner	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M21Kner.140721	CAGCATGTGTTG	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	right_back_of_knees	Human Skin	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	right back of knees	FFO92CG	36	mimarks-survey	NA	FMA:Right popliteal fossa	NA	NA	0	UBERON:zone of skin of knee	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M2	sample337	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M21Kner	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	36	M2	right popliteal fossa	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFO92CG	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M21Kner	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M31Kner.140321	CGCTTATCGAGA	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	right_back_of_knees	Human Skin	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	right back of knees	FFLHOYS	33	mimarks-survey	NA	FMA:Right popliteal fossa	NA	NA	0	UBERON:zone of skin of knee	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M3	sample339	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M31Kner	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	M3	right popliteal fossa	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M31Kner	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M41Kner.140735	CTGGCTGTATGA	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	right_back_of_knees	Human Skin	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	right back of knees	FFO92CG	31	mimarks-survey	NA	FMA:Right popliteal fossa	NA	NA	0	UBERON:zone of skin of knee	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M4	sample341	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M41Kner	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	31	M4	right popliteal fossa	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFO92CG	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M41Kner	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M42Kner.140502	GACTAGACCAGC	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	right_back_of_knees	Human Skin	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	right back of knees	FFO92CG	31	mimarks-survey	NA	FMA:Right popliteal fossa	NA	NA	0	UBERON:zone of skin of knee	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M4	sample342	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M42Kner	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	31	M4	right popliteal fossa	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFO92CG	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M42Kner	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
F11Forr.140422	AGAGTAGCTAAG	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	right_forearm	Human Skin	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	right forearm	FFLHOYS	33	mimarks-survey	NA	FMA:Surface of right arm	NA	NA	0	UBERON:skin of forearm	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F1	sample343	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F11Forr	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	F1	right volar forearm	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F11Forr	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
F12Forr.140302	AGTCCATAGCTG	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	right_forearm	Human Skin	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	right forearm	FFLHOYS	33	mimarks-survey	NA	FMA:Surface of right arm	NA	NA	0	UBERON:skin of forearm	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F1	sample344	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F12Forr	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	F1	right volar forearm	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F12Forr	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
F21Forr.140353	ATCGCTCGAGGA	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	right_forearm	Human Skin	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	right forearm	FFLHOYS	31	mimarks-survey	NA	FMA:Surface of right arm	NA	NA	0	UBERON:skin of forearm	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F2	sample345	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F21Forr	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	31	F2	right volar forearm	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F21Forr	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
F22Forr.140732	CAAGATCGACTC	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	right_forearm	Human Skin	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	right forearm	FFLHOYS	31	mimarks-survey	NA	FMA:Surface of right arm	NA	NA	0	UBERON:skin of forearm	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F2	sample346	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F22Forr	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	31	F2	right volar forearm	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F22Forr	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
F31Forr.140586	GATGTCGTGTCA	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	right_forearm	Human Skin	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	right forearm	FFO92CG	33	mimarks-survey	NA	FMA:Surface of right arm	NA	NA	0	UBERON:skin of forearm	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F3	sample347	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F31Forr	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	F3	right volar forearm	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFO92CG	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F31Forr	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
F32Forr.140528	GCGTTACACACA	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	right_forearm	Human Skin	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	right forearm	FFO92CG	33	mimarks-survey	NA	FMA:Surface of right arm	NA	NA	0	UBERON:skin of forearm	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F3	sample348	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F32Forr	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	F3	right volar forearm	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFO92CG	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F32Forr	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M12Forr.140849	ACGGATCGTCAG	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	right_forearm	Human Skin	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	right forearm	FFLHOYS	36	mimarks-survey	NA	FMA:Surface of right arm	NA	NA	0	UBERON:skin of forearm	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M1	sample350	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M12Forr	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	36	M1	right volar forearm	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M12Forr	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M21Forr.140361	CAGTCGAAGCTG	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	right_forearm	Human Skin	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	right forearm	FFO92CG	36	mimarks-survey	NA	FMA:Surface of right arm	NA	NA	0	UBERON:skin of forearm	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M2	sample351	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M21Forr	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	36	M2	right volar forearm	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFO92CG	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M21Forr	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M22Forr.140708	CGAATCGACACT	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	right_forearm	Human Skin	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	right forearm	FFO92CG	36	mimarks-survey	NA	FMA:Surface of right arm	NA	NA	0	UBERON:skin of forearm	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M2	sample352	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M22Forr	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	36	M2	right volar forearm	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFO92CG	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M22Forr	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M32Forr.140802	CTATCTAGCGAG	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	right_forearm	Human Skin	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	right forearm	FFLHOYS	33	mimarks-survey	NA	FMA:Surface of right arm	NA	NA	0	UBERON:skin of forearm	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M3	sample354	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M32Forr	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	M3	right volar forearm	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M32Forr	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M41Forr.140728	CTTAGCACATCA	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	right_forearm	Human Skin	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	right forearm	FFO92CG	31	mimarks-survey	NA	FMA:Surface of right arm	NA	NA	0	UBERON:skin of forearm	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M4	sample355	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M41Forr	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	31	M4	right volar forearm	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFO92CG	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M41Forr	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M42Forr.140521	GACTTCAGTGTG	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	right_forearm	Human Skin	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	right forearm	FFO92CG	31	mimarks-survey	NA	FMA:Surface of right arm	NA	NA	0	UBERON:skin of forearm	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M4	sample356	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M42Forr	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	31	M4	right volar forearm	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFO92CG	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M42Forr	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
F11Indr.140809	AGATCTCTGCAT	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	right_index_finger	Human Skin	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	right index finger	FFLHOYS	33	mimarks-survey	NA	FMA:Right index finger	NA	NA	0	UBERON:skin of finger	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F1	sample357	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F11Indr	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	F1	right palmar index finger	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F11Indr	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
F12Indr.140556	AGTGCGATGCGT	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	right_index_finger	Human Skin	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	right index finger	FFLHOYS	33	mimarks-survey	NA	FMA:Right index finger	NA	NA	0	UBERON:skin of finger	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F1	sample358	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F12Indr	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	F1	right palmar index finger	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F12Indr	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
F21Indr.140440	ATCTGAGCTGGT	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	right_index_finger	Human Skin	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	right index finger	FFLHOYS	31	mimarks-survey	NA	FMA:Right index finger	NA	NA	0	UBERON:skin of finger	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F2	sample359	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F21Indr	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	31	F2	right palmar index finger	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F21Indr	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
F22Indr.140640	CACAGTGGACGT	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	right_index_finger	Human Skin	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	right index finger	FFLHOYS	31	mimarks-survey	NA	FMA:Right index finger	NA	NA	0	UBERON:skin of finger	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F2	sample360	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F22Indr	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	31	F2	right palmar index finger	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F22Indr	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
F31Indr.140675	GCACATCGAGCA	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	right_index_finger	Human Skin	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	right index finger	FFO92CG	33	mimarks-survey	NA	FMA:Right index finger	NA	NA	0	UBERON:skin of finger	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F3	sample361	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F31Indr	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	F3	right palmar index finger	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFO92CG	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F31Indr	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
F32Indr.140459	GCTATCACGAGT	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	right_index_finger	Human Skin	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	right index finger	FFO92CG	33	mimarks-survey	NA	FMA:Right index finger	NA	NA	0	UBERON:skin of finger	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F3	sample362	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F32Indr	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	F3	right palmar index finger	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFO92CG	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F32Indr	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M11Indr.140633	ACAGAGTCGGCT	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	right_index_finger	Human Skin	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	right index finger	FFLHOYS	36	mimarks-survey	NA	FMA:Right index finger	NA	NA	0	UBERON:skin of finger	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M1	sample363	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M11Indr	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	36	M1	right palmar index finger	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M11Indr	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M12Indr.140762	ACGTGCCGTAGA	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	right_index_finger	Human Skin	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	right index finger	FFLHOYS	36	mimarks-survey	NA	FMA:Right index finger	NA	NA	0	UBERON:skin of finger	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M1	sample364	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M12Indr	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	36	M1	right palmar index finger	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M12Indr	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M21Indr.140425	CATAGACGTTCG	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	right_index_finger	Human Skin	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	right index finger	FFO92CG	36	mimarks-survey	NA	FMA:Right index finger	NA	NA	0	UBERON:skin of finger	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M2	sample365	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M21Indr	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	36	M2	right palmar index finger	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFO92CG	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M21Indr	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M22Indr.140471	CGAGCAGCACAT	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	right_index_finger	Human Skin	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	right index finger	FFO92CG	36	mimarks-survey	NA	FMA:Right index finger	NA	NA	0	UBERON:skin of finger	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M2	sample366	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M22Indr	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	36	M2	right palmar index finger	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFO92CG	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M22Indr	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M31Indr.140588	CGTCAGACGGAT	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	right_index_finger	Human Skin	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	right index finger	FFLHOYS	33	mimarks-survey	NA	FMA:Right index finger	NA	NA	0	UBERON:skin of finger	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M3	sample367	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M31Indr	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	M3	right palmar index finger	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M31Indr	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M32Indr.140748	CTCCACATGAGA	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	right_index_finger	Human Skin	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	right index finger	FFLHOYS	33	mimarks-survey	NA	FMA:Right index finger	NA	NA	0	UBERON:skin of finger	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M3	sample368	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M32Indr	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	M3	right palmar index finger	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M32Indr	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M41Indr.140781	GAAGAGTGATCA	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	right_index_finger	Human Skin	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	right index finger	FFO92CG	31	mimarks-survey	NA	FMA:Right index finger	NA	NA	0	UBERON:skin of finger	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M4	sample369	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M41Indr	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	31	M4	right palmar index finger	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFO92CG	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M41Indr	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M42Indr.140519	GAGATGCCGACT	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	right_index_finger	Human Skin	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	right index finger	FFO92CG	31	mimarks-survey	NA	FMA:Right index finger	NA	NA	0	UBERON:skin of finger	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M4	sample370	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M42Indr	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	31	M4	right palmar index finger	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFO92CG	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M42Indr	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
F11Nstr.140303	AGCATATGAGAG	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	right_nostril	Human Skin	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	right nostril	FFLHOYS	33	mimarks-survey	NA	FMA:Nares	NA	NA	0	UBERON:nostril	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F1	sample371	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:nostril	UBERON:mucus	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F11Nstr	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	F1	right naris	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human nose metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	646099	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F11Nstr	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human nose metagenome
F12Nstr.140310	AGTTCTACGTCA	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	right_nostril	Human Skin	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	right nostril	FFLHOYS	33	mimarks-survey	NA	FMA:Nares	NA	NA	0	UBERON:nostril	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F1	sample372	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:nostril	UBERON:mucus	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F12Nstr	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	F1	right naris	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human nose metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	646099	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F12Nstr	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human nose metagenome
F21Nstr.140791	TCATCGCGATAT	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	right_nostril	Human Skin	source	NA	n	11/14/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	right nostril	FKB0RMH	31	mimarks-survey	NA	FMA:Nares	NA	NA	0	UBERON:nostril	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F2	sample373	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:nostril	UBERON:mucus	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F21Nstr	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	31	F2	right naris	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human nose metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	646099	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F21Nstr	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human nose metagenome
F22Nstr.140671	CACGTGACATGT	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	right_nostril	Human Skin	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	right nostril	FFLHOYS	31	mimarks-survey	NA	FMA:Nares	NA	NA	0	UBERON:nostril	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F2	sample374	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:nostril	UBERON:mucus	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F22Nstr	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	31	F2	right naris	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human nose metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	646099	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F22Nstr	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human nose metagenome
F31Nstr.140608	GCAGGATAGATA	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	right_nostril	Human Skin	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	right nostril	FFO92CG	33	mimarks-survey	NA	FMA:Nares	NA	NA	0	UBERON:nostril	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F3	sample375	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:nostril	UBERON:mucus	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F31Nstr	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	F3	right naris	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human nose metagenome	NA	ENVO:human-associated habitat	FLX	FFO92CG	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	646099	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F31Nstr	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human nose metagenome
F32Nstr.140767	GCTGGTATCTGA	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	right_nostril	Human Skin	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	right nostril	FFO92CG	33	mimarks-survey	NA	FMA:Nares	NA	NA	0	UBERON:nostril	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F3	sample376	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:nostril	UBERON:mucus	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F32Nstr	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	F3	right naris	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human nose metagenome	NA	ENVO:human-associated habitat	FLX	FFO92CG	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	646099	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F32Nstr	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human nose metagenome
M11Nstr.140514	ACATCACTTAGC	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	right_nostril	Human Skin	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	right nostril	FFLHOYS	36	mimarks-survey	NA	FMA:Nares	NA	NA	0	UBERON:nostril	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M1	sample377	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:nostril	UBERON:mucus	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M11Nstr	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	36	M1	right naris	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human nose metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	646099	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M11Nstr	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human nose metagenome
M12Nstr.140690	ACTATTGTCACG	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	right_nostril	Human Skin	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	right nostril	FFLHOYS	36	mimarks-survey	NA	FMA:Nares	NA	NA	0	UBERON:nostril	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M1	sample378	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:nostril	UBERON:mucus	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M12Nstr	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	36	M1	right naris	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human nose metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	646099	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M12Nstr	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human nose metagenome
M21Nstr.140593	CATCATGAGGCT	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	right_nostril	Human Skin	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	right nostril	FFO92CG	36	mimarks-survey	NA	FMA:Nares	NA	NA	0	UBERON:nostril	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M2	sample379	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:nostril	UBERON:mucus	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M21Nstr	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	36	M2	right naris	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human nose metagenome	NA	ENVO:human-associated habitat	FLX	FFO92CG	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	646099	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M21Nstr	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human nose metagenome
M22Nstr.140393	CGATATTCATCG	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	right_nostril	Human Skin	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	right nostril	FFO92CG	36	mimarks-survey	NA	FMA:Nares	NA	NA	0	UBERON:nostril	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M2	sample380	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:nostril	UBERON:mucus	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M22Nstr	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	36	M2	right naris	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human nose metagenome	NA	ENVO:human-associated habitat	FLX	FFO92CG	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	646099	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M22Nstr	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human nose metagenome
M31Nstr.140461	CGTGTACATCAG	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	right_nostril	Human Skin	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	right nostril	FFLHOYS	33	mimarks-survey	NA	FMA:Nares	NA	NA	0	UBERON:nostril	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M3	sample381	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:nostril	UBERON:mucus	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M31Nstr	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	M3	right naris	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human nose metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	646099	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M31Nstr	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human nose metagenome
M32Nstr.140494	CTCGTGGAGTAG	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	right_nostril	Human Skin	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	right nostril	FFLHOYS	33	mimarks-survey	NA	FMA:Nares	NA	NA	0	UBERON:nostril	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M3	sample382	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:nostril	UBERON:mucus	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M32Nstr	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	M3	right naris	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human nose metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	646099	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M32Nstr	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human nose metagenome
M41Nstr.140549	GACAGCGTTGAC	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	right_nostril	Human Skin	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	right nostril	FFO92CG	31	mimarks-survey	NA	FMA:Nares	NA	NA	0	UBERON:nostril	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M4	sample383	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:nostril	UBERON:mucus	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M41Nstr	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	31	M4	right naris	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human nose metagenome	NA	ENVO:human-associated habitat	FLX	FFO92CG	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	646099	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M41Nstr	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human nose metagenome
M42Nstr.140784	GAGTAGCTCGTG	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	right_nostril	Human Skin	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	right nostril	FFO92CG	31	mimarks-survey	NA	FMA:Nares	NA	NA	0	UBERON:nostril	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M4	sample384	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:nostril	UBERON:mucus	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M42Nstr	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	31	M4	right naris	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human nose metagenome	NA	ENVO:human-associated habitat	FLX	FFO92CG	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	646099	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M42Nstr	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human nose metagenome
F11Pinr.140867	TGCGCGAATACT	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	right_outer_ear	Human Skin	source	NA	n	11/14/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	right outer ear	FKB0RMH	33	mimarks-survey	NA	FMA:Surface of right pinna	NA	NA	0	UBERON:zone of skin of outer ear	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F1	sample385	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F11Pinr	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	F1	right lateral pinna	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F11Pinr	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
F12Pinr.140498	ATACGTCTTCGA	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	right_outer_ear	Human Skin	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	right outer ear	FFLHOYS	33	mimarks-survey	NA	FMA:Surface of right pinna	NA	NA	0	UBERON:zone of skin of outer ear	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F1	sample386	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F12Pinr	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	F1	right lateral pinna	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F12Pinr	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
F21Pinr.140773	ATGCGTAGTGCG	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	right_outer_ear	Human Skin	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	right outer ear	FFLHOYS	31	mimarks-survey	NA	FMA:Surface of right pinna	NA	NA	0	UBERON:zone of skin of outer ear	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F2	sample387	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F21Pinr	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	31	F2	right lateral pinna	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F21Pinr	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
F22Pinr.140355	CACTGGTATATC	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	right_outer_ear	Human Skin	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	right outer ear	FFLHOYS	31	mimarks-survey	NA	FMA:Surface of right pinna	NA	NA	0	UBERON:zone of skin of outer ear	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F2	sample388	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F22Pinr	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	31	F2	right lateral pinna	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F22Pinr	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
F31Pinr.140799	GCATAGTAGCCG	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	right_outer_ear	Human Skin	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	right outer ear	FFO92CG	33	mimarks-survey	NA	FMA:Surface of right pinna	NA	NA	0	UBERON:zone of skin of outer ear	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F3	sample389	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F31Pinr	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	F3	right lateral pinna	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFO92CG	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F31Pinr	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
F32Pinr.140328	GCTTCATAGTGT	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	right_outer_ear	Human Skin	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	right outer ear	FFO92CG	33	mimarks-survey	NA	FMA:Surface of right pinna	NA	NA	0	UBERON:zone of skin of outer ear	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F3	sample390	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F32Pinr	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	F3	right lateral pinna	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFO92CG	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F32Pinr	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M11Pinr.140546	ACCACATACATC	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	right_outer_ear	Human Skin	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	right outer ear	FFLHOYS	36	mimarks-survey	NA	FMA:Surface of right pinna	NA	NA	0	UBERON:zone of skin of outer ear	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M1	sample391	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M11Pinr	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	36	M1	right lateral pinna	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M11Pinr	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M12Pinr.140598	ACTCGCACAGGA	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	right_outer_ear	Human Skin	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	right outer ear	FFLHOYS	36	mimarks-survey	NA	FMA:Surface of right pinna	NA	NA	0	UBERON:zone of skin of outer ear	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M1	sample392	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M12Pinr	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	36	M1	right lateral pinna	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M12Pinr	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M21Pinr.140516	CATGCAGACTGT	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	right_outer_ear	Human Skin	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	right outer ear	FFO92CG	36	mimarks-survey	NA	FMA:Surface of right pinna	NA	NA	0	UBERON:zone of skin of outer ear	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M2	sample393	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M21Pinr	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	36	M2	right lateral pinna	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFO92CG	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M21Pinr	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M22Pinr.140692	CGCACATGTTAT	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	right_outer_ear	Human Skin	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	right outer ear	FFO92CG	36	mimarks-survey	NA	FMA:Surface of right pinna	NA	NA	0	UBERON:zone of skin of outer ear	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M2	sample394	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M22Pinr	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	36	M2	right lateral pinna	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFO92CG	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M22Pinr	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M31Pinr.140627	CGTTCGCATAGA	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	right_outer_ear	Human Skin	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	right outer ear	FFLHOYS	33	mimarks-survey	NA	FMA:Surface of right pinna	NA	NA	0	UBERON:zone of skin of outer ear	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M3	sample395	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M31Pinr	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	M3	right lateral pinna	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M31Pinr	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M32Pinr.140389	CTGAACGCTAGT	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	right_outer_ear	Human Skin	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	right outer ear	FFLHOYS	33	mimarks-survey	NA	FMA:Surface of right pinna	NA	NA	0	UBERON:zone of skin of outer ear	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M3	sample396	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M32Pinr	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	M3	right lateral pinna	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M32Pinr	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M41Pinr.140509	GACCACTACGAT	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	right_outer_ear	Human Skin	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	right outer ear	FFO92CG	31	mimarks-survey	NA	FMA:Surface of right pinna	NA	NA	0	UBERON:zone of skin of outer ear	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M4	sample397	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M41Pinr	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	31	M4	right lateral pinna	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFO92CG	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M41Pinr	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M42Pinr.140457	GAGTGGTAGAGA	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	right_outer_ear	Human Skin	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	right outer ear	FFO92CG	31	mimarks-survey	NA	FMA:Surface of right pinna	NA	NA	0	UBERON:zone of skin of outer ear	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M4	sample398	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M42Pinr	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	31	M4	right lateral pinna	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFO92CG	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M42Pinr	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
F11Ewxr.140797	AGACGTGCACTG	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	right_outer_ear_canal_earwax	Human Skin	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	right outer ear canal/earwax	FFLHOYS	33	mimarks-survey	NA	FMA:Right external auditory canal	NA	NA	0	UBERON:zone of skin of outer ear	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F1	sample399	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:external auditory canal	UBERON:cerumen	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F11Ewxr	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	F1	right external auditory canal	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	646099	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F11Ewxr	ENVO:ear wax	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human metagenome
F12Ewxr.140726	AGTACGCTCGAG	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	right_outer_ear_canal_earwax	Human Skin	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	right outer ear canal/earwax	FFLHOYS	33	mimarks-survey	NA	FMA:Right external auditory canal	NA	NA	0	UBERON:zone of skin of outer ear	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F1	sample400	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:external auditory canal	UBERON:cerumen	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F12Ewxr	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	F1	right external auditory canal	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	646099	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F12Ewxr	ENVO:ear wax	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human metagenome
F21Ewxr.140299	ATCCGATCACAG	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	right_outer_ear_canal_earwax	Human Skin	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	right outer ear canal/earwax	FFLHOYS	31	mimarks-survey	NA	FMA:Right external auditory canal	NA	NA	0	UBERON:zone of skin of outer ear	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F2	sample401	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:external auditory canal	UBERON:cerumen	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F21Ewxr	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	31	F2	right external auditory canal	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	646099	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F21Ewxr	ENVO:ear wax	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human metagenome
F22Ewxr.140747	ATTCTGTGAGCG	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	right_outer_ear_canal_earwax	Human Skin	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	right outer ear canal/earwax	FFLHOYS	31	mimarks-survey	NA	FMA:Right external auditory canal	NA	NA	0	UBERON:zone of skin of outer ear	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F2	sample402	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:external auditory canal	UBERON:cerumen	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F22Ewxr	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	31	F2	right external auditory canal	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	646099	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F22Ewxr	ENVO:ear wax	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human metagenome
F31Ewxr.140289	GATCTCATAGGC	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	right_outer_ear_canal_earwax	Human Skin	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	right outer ear canal/earwax	FFO92CG	33	mimarks-survey	NA	FMA:Right external auditory canal	NA	NA	0	UBERON:zone of skin of outer ear	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F3	sample403	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:external auditory canal	UBERON:cerumen	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F31Ewxr	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	F3	right external auditory canal	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human metagenome	NA	ENVO:human-associated habitat	FLX	FFO92CG	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	646099	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F31Ewxr	ENVO:ear wax	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human metagenome
F32Ewxr.140535	GCGATATATCGC	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	right_outer_ear_canal_earwax	Human Skin	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	right outer ear canal/earwax	FFO92CG	33	mimarks-survey	NA	FMA:Right external auditory canal	NA	NA	0	UBERON:zone of skin of outer ear	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F3	sample404	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:external auditory canal	UBERON:cerumen	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F32Ewxr	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	F3	right external auditory canal	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human metagenome	NA	ENVO:human-associated habitat	FLX	FFO92CG	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	646099	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F32Ewxr	ENVO:ear wax	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human metagenome
M12Ewxr.140434	ACGCGATACTGG	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	right_outer_ear_canal_earwax	Human Skin	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	right outer ear canal/earwax	FFLHOYS	36	mimarks-survey	NA	FMA:Right external auditory canal	NA	NA	0	UBERON:zone of skin of outer ear	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M1	sample405	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:external auditory canal	UBERON:cerumen	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M12Ewxr	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	36	M1	right external auditory canal	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	646099	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M12Ewxr	ENVO:ear wax	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human metagenome
M21Ewxr.140654	CAGCTAGAACGC	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	right_outer_ear_canal_earwax	Human Skin	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	right outer ear canal/earwax	FFO92CG	36	mimarks-survey	NA	FMA:Right external auditory canal	NA	NA	0	UBERON:zone of skin of outer ear	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M2	sample406	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:external auditory canal	UBERON:cerumen	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M21Ewxr	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	36	M2	right external auditory canal	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human metagenome	NA	ENVO:human-associated habitat	FLX	FFO92CG	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	646099	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M21Ewxr	ENVO:ear wax	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human metagenome
M22Ewxr.140787	CCGATGTCAGAT	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	right_outer_ear_canal_earwax	Human Skin	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	right outer ear canal/earwax	FFO92CG	36	mimarks-survey	NA	FMA:Right external auditory canal	NA	NA	0	UBERON:zone of skin of outer ear	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M2	sample407	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:external auditory canal	UBERON:cerumen	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M22Ewxr	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	36	M2	right external auditory canal	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human metagenome	NA	ENVO:human-associated habitat	FLX	FFO92CG	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	646099	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M22Ewxr	ENVO:ear wax	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human metagenome
M31Ewxr.140754	CGGCGATGTACA	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	right_outer_ear_canal_earwax	Human Skin	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	right outer ear canal/earwax	FFLHOYS	33	mimarks-survey	NA	FMA:Right external auditory canal	NA	NA	0	UBERON:zone of skin of outer ear	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M3	sample408	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:external auditory canal	UBERON:cerumen	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M31Ewxr	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	M3	right external auditory canal	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	646099	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M31Ewxr	ENVO:ear wax	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human metagenome
M32Ewxr.140697	CTAGGTCACTAG	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	right_outer_ear_canal_earwax	Human Skin	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	right outer ear canal/earwax	FFLHOYS	33	mimarks-survey	NA	FMA:Right external auditory canal	NA	NA	0	UBERON:zone of skin of outer ear	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M3	sample409	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:external auditory canal	UBERON:cerumen	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M32Ewxr	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	M3	right external auditory canal	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	646099	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M32Ewxr	ENVO:ear wax	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human metagenome
M41Ewxr.140794	CTGTCTCTCCTA	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	right_outer_ear_canal_earwax	Human Skin	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	right outer ear canal/earwax	FFO92CG	31	mimarks-survey	NA	FMA:Right external auditory canal	NA	NA	0	UBERON:zone of skin of outer ear	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M4	sample410	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:external auditory canal	UBERON:cerumen	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M41Ewxr	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	31	M4	right external auditory canal	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human metagenome	NA	ENVO:human-associated habitat	FLX	FFO92CG	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	646099	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M41Ewxr	ENVO:ear wax	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human metagenome
M42Ewxr.140604	GACTCGAATCGT	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	right_outer_ear_canal_earwax	Human Skin	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	right outer ear canal/earwax	FFO92CG	31	mimarks-survey	NA	FMA:Right external auditory canal	NA	NA	0	UBERON:zone of skin of outer ear	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M4	sample411	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:external auditory canal	UBERON:cerumen	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M42Ewxr	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	31	M4	right external auditory canal	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human metagenome	NA	ENVO:human-associated habitat	FLX	FFO92CG	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	646099	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M42Ewxr	ENVO:ear wax	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human metagenome
F11Plmr.140656	AGCGACTGTGCA	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	right_palm	Human Skin	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	right palm	FFLHOYS	33	mimarks-survey	NA	FMA:Right palm	NA	NA	0	UBERON:zone of skin of hand	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F1	sample412	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F11Plmr	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	F1	right palm	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F11Plmr	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
F12Plmr.140839	ATACACGTGGCG	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	right_palm	Human Skin	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	right palm	FFLHOYS	33	mimarks-survey	NA	FMA:Right palm	NA	NA	0	UBERON:zone of skin of hand	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F1	sample413	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F12Plmr	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	F1	right palm	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F12Plmr	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
F13Plmr.140454	GACTCACTCAAT	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	right_palm	Human Skin	source	NA	n	11/14/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	right palm	FKB0RMH	33	mimarks-survey	NA	FMA:Right palm	NA	NA	0	UBERON:zone of skin of hand	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F1	sample414	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F13Plmr	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	F1	right palm	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F13Plmr	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
F14Plmr.140786	GAGCATTCTCTA	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	right_palm	Human Skin	source	NA	n	11/14/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	right palm	FKB0RMH	33	mimarks-survey	NA	FMA:Right palm	NA	NA	0	UBERON:zone of skin of hand	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F1	sample415	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F14Plmr	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	F1	right palm	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F14Plmr	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
F21Plmr.140745	ATGCAGCTCAGT	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	right_palm	Human Skin	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	right palm	FFLHOYS	31	mimarks-survey	NA	FMA:Right palm	NA	NA	0	UBERON:zone of skin of hand	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F2	sample416	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F21Plmr	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	31	F2	right palm	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F21Plmr	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
F22Plmr.140357	CACTCAACAGAC	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	right_palm	Human Skin	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	right palm	FFLHOYS	31	mimarks-survey	NA	FMA:Right palm	NA	NA	0	UBERON:zone of skin of hand	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F2	sample417	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F22Plmr	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	31	F2	right palm	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F22Plmr	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
F23Plmr.140435	GCATATAGTCTC	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	right_palm	Human Skin	source	NA	n	11/14/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	right palm	FKB0RMH	31	mimarks-survey	NA	FMA:Right palm	NA	NA	0	UBERON:zone of skin of hand	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F2	sample418	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F23Plmr	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	31	F2	right palm	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F23Plmr	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
F24Plmr.140433	GCGTATCTTGAT	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	right_palm	Human Skin	source	NA	n	11/14/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	right palm	FKB0RMH	31	mimarks-survey	NA	FMA:Right palm	NA	NA	0	UBERON:zone of skin of hand	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F2	sample419	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F24Plmr	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	31	F2	right palm	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F24Plmr	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
F31Plmr.140385	GCAGTATCACTG	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	right_palm	Human Skin	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	right palm	FFO92CG	33	mimarks-survey	NA	FMA:Right palm	NA	NA	0	UBERON:zone of skin of hand	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F3	sample420	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F31Plmr	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	F3	right palm	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFO92CG	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F31Plmr	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
F32Plmr.140529	GCTGTGTAGGAC	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	right_palm	Human Skin	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	right palm	FFO92CG	33	mimarks-survey	NA	FMA:Right palm	NA	NA	0	UBERON:zone of skin of hand	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F3	sample421	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F32Plmr	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	F3	right palm	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFO92CG	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F32Plmr	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
F33Plmr.140674	GTAGTGTCTAGC	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	right_palm	Human Skin	source	NA	n	11/14/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	right palm	FKB0RMH	33	mimarks-survey	NA	FMA:Right palm	NA	NA	0	UBERON:zone of skin of hand	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F3	sample422	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F33Plmr	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	F3	right palm	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F33Plmr	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
F34Plmr.140416	GTCGCTGTCTTC	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	right_palm	Human Skin	source	NA	n	11/14/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	right palm	FKB0RMH	33	mimarks-survey	NA	FMA:Right palm	NA	NA	0	UBERON:zone of skin of hand	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F3	sample423	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F34Plmr	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	F3	right palm	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F34Plmr	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M11Plmr.140866	ACATGTCACGTG	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	right_palm	Human Skin	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	right palm	FFLHOYS	36	mimarks-survey	NA	FMA:Right palm	NA	NA	0	UBERON:zone of skin of hand	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M1	sample424	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M11Plmr	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	36	M1	right palm	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M11Plmr	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M12Plmr.140584	ACTCAGATACTC	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	right_palm	Human Skin	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	right palm	FFLHOYS	36	mimarks-survey	NA	FMA:Right palm	NA	NA	0	UBERON:zone of skin of hand	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M1	sample425	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M12Plmr	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	36	M1	right palm	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M12Plmr	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M13Plmr.140301	CTGTTCGTAGAG	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	right_palm	Human Skin	source	NA	n	11/14/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	right palm	FKB0RMH	36	mimarks-survey	NA	FMA:Right palm	NA	NA	0	UBERON:zone of skin of hand	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M1	sample426	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M13Plmr	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	36	M1	right palm	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M13Plmr	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M14Plmr.140579	GACAGGAGATAG	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	right_palm	Human Skin	source	NA	n	11/14/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	right palm	FKB0RMH	36	mimarks-survey	NA	FMA:Right palm	NA	NA	0	UBERON:zone of skin of hand	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M1	sample427	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M14Plmr	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	36	M1	right palm	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M14Plmr	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M21Plmr.140842	CATCTGTAGCGA	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	right_palm	Human Skin	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	right palm	FFO92CG	36	mimarks-survey	NA	FMA:Right palm	NA	NA	0	UBERON:zone of skin of hand	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M2	sample428	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M21Plmr	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	36	M2	right palm	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFO92CG	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M21Plmr	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M22Plmr.140574	CGATGCACCAGA	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	right_palm	Human Skin	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	right palm	FFO92CG	36	mimarks-survey	NA	FMA:Right palm	NA	NA	0	UBERON:zone of skin of hand	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M2	sample429	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M22Plmr	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	36	M2	right palm	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFO92CG	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M22Plmr	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M23Plmr.140648	GATCAGAAGATG	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	right_palm	Human Skin	source	NA	n	11/14/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	right palm	FKB0RMH	36	mimarks-survey	NA	FMA:Right palm	NA	NA	0	UBERON:zone of skin of hand	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M2	sample430	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M23Plmr	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	36	M2	right palm	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M23Plmr	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M24Plmr.140833	GCAATAGCTGCT	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	right_palm	Human Skin	source	NA	n	11/14/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	right palm	FKB0RMH	36	mimarks-survey	NA	FMA:Right palm	NA	NA	0	UBERON:zone of skin of hand	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M2	sample431	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M24Plmr	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	36	M2	right palm	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M24Plmr	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M31Plmr.140505	CGTTACTAGAGC	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	right_palm	Human Skin	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	right palm	FFLHOYS	33	mimarks-survey	NA	FMA:Right palm	NA	NA	0	UBERON:zone of skin of hand	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M3	sample432	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M31Plmr	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	M3	right palm	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M31Plmr	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M32Plmr.140307	CTCTGAAGTCTA	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	right_palm	Human Skin	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	right palm	FFLHOYS	33	mimarks-survey	NA	FMA:Right palm	NA	NA	0	UBERON:zone of skin of hand	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M3	sample433	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M32Plmr	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	M3	right palm	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M32Plmr	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M33Plmr.140723	GCTGTAGTATGC	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	right_palm	Human Skin	source	NA	n	11/14/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	right palm	FKB0RMH	33	mimarks-survey	NA	FMA:Right palm	NA	NA	0	UBERON:zone of skin of hand	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M3	sample434	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M33Plmr	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	M3	right palm	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M33Plmr	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M34Plmr.140826	GGTCACTGACAG	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	right_palm	Human Skin	source	NA	n	11/14/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	right palm	FKB0RMH	33	mimarks-survey	NA	FMA:Right palm	NA	NA	0	UBERON:zone of skin of hand	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M3	sample435	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M34Plmr	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	M3	right palm	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M34Plmr	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M41Plmr.140782	GACAGTTACTGC	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	right_palm	Human Skin	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	right palm	FFO92CG	31	mimarks-survey	NA	FMA:Right palm	NA	NA	0	UBERON:zone of skin of hand	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M4	sample436	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M41Plmr	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	31	M4	right palm	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFO92CG	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M41Plmr	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M42Plmr.140511	GAGTCTGAGTCT	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	right_palm	Human Skin	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	right palm	FFO92CG	31	mimarks-survey	NA	FMA:Right palm	NA	NA	0	UBERON:zone of skin of hand	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M4	sample437	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M42Plmr	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	31	M4	right palm	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFO92CG	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M42Plmr	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M43Plmr.140716	GTGAGGTCGCTA	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	right_palm	Human Skin	source	NA	n	11/14/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	right palm	FKB0RMH	31	mimarks-survey	NA	FMA:Right palm	NA	NA	0	UBERON:zone of skin of hand	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M4	sample438	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M43Plmr	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	31	M4	right palm	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M43Plmr	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M44Plmr.140868	GTTGACGACAGC	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	right_palm	Human Skin	source	NA	n	11/14/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	right palm	FKB0RMH	31	mimarks-survey	NA	FMA:Right palm	NA	NA	0	UBERON:zone of skin of hand	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M4	sample439	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M44Plmr	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	31	M4	right palm	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M44Plmr	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M53Plmr.140520	TACGCGCTGAGA	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	right_palm	Human Skin	source	NA	n	11/14/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	right palm	FKB0RMH	60	mimarks-survey	NA	FMA:Right palm	NA	NA	0	UBERON:zone of skin of hand	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M5	sample440	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M53Plmr	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	60	M5	right palm	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M53Plmr	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M54Plmr.140626	TAGATCCTCGAT	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	right_palm	Human Skin	source	NA	n	11/14/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	right palm	FKB0RMH	60	mimarks-survey	NA	FMA:Right palm	NA	NA	0	UBERON:zone of skin of hand	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M5	sample441	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M54Plmr	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	60	M5	right palm	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M54Plmr	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M63Plmr.140815	TAGTTGCGAGTC	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	right_palm	Human Skin	source	NA	n	11/14/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	right palm	FKB0RMH	NA	mimarks-survey	NA	FMA:Right palm	NA	NA	0	UBERON:zone of skin of hand	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M6	sample442	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M63Plmr	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	NA	M6	right palm	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M63Plmr	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M64Plmr.140356	TCACGATTAGCG	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	right_palm	Human Skin	source	NA	n	11/14/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	right palm	FKB0RMH	NA	mimarks-survey	NA	FMA:Right palm	NA	NA	0	UBERON:zone of skin of hand	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M6	sample443	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M64Plmr	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	NA	M6	right palm	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M64Plmr	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
F11Fotr.140431	AGCTATCCACGA	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	right_sole_of_foot	Human Skin	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	right sole of foot	FFLHOYS	33	mimarks-survey	NA	FMA:Right foot surface	NA	NA	0	UBERON:zone of skin of foot	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F1	sample444	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F11Fotr	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	F1	right plantar foot	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F11Fotr	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
F12Fotr.140597	ATACTCACTCAG	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	right_sole_of_foot	Human Skin	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	right sole of foot	FFLHOYS	33	mimarks-survey	NA	FMA:Right foot surface	NA	NA	0	UBERON:zone of skin of foot	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F1	sample445	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F12Fotr	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	F1	right plantar foot	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F12Fotr	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
F13Fotr.140798	GACTTCAGTGTG	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	right_sole_of_foot	Human Skin	source	NA	n	11/14/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	right sole of foot	FKB0RMH	33	mimarks-survey	NA	FMA:Right foot surface	NA	NA	0	UBERON:zone of skin of foot	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F1	sample446	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F13Fotr	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	F1	right plantar foot	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F13Fotr	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
F14Fotr.140665	GAGTCTGAGTCT	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	right_sole_of_foot	Human Skin	source	NA	n	11/14/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	right sole of foot	FKB0RMH	33	mimarks-survey	NA	FMA:Right foot surface	NA	NA	0	UBERON:zone of skin of foot	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F1	sample447	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F14Fotr	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	F1	right plantar foot	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F14Fotr	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
F21Fotr.140485	ATGGCAGCTCTA	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	right_sole_of_foot	Human Skin	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	right sole of foot	FFLHOYS	31	mimarks-survey	NA	FMA:Right foot surface	NA	NA	0	UBERON:zone of skin of foot	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F2	sample448	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F21Fotr	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	31	F2	right plantar foot	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F21Fotr	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
F22Fotr.140698	CAGACATTGCGT	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	right_sole_of_foot	Human Skin	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	right sole of foot	FFLHOYS	31	mimarks-survey	NA	FMA:Right foot surface	NA	NA	0	UBERON:zone of skin of foot	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F2	sample449	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F22Fotr	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	31	F2	right plantar foot	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F22Fotr	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
F23Fotr.140680	GCCAGAGTCGTA	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	right_sole_of_foot	Human Skin	source	NA	n	11/14/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	right sole of foot	FKB0RMH	31	mimarks-survey	NA	FMA:Right foot surface	NA	NA	0	UBERON:zone of skin of foot	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F2	sample450	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F23Fotr	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	31	F2	right plantar foot	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F23Fotr	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
F24Fotr.140273	GCTATCACGAGT	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	right_sole_of_foot	Human Skin	source	NA	n	11/14/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	right sole of foot	FKB0RMH	31	mimarks-survey	NA	FMA:Right foot surface	NA	NA	0	UBERON:zone of skin of foot	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F2	sample451	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F24Fotr	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	31	F2	right plantar foot	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F24Fotr	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
F31Fotr.140874	GCATCGTCAACA	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	right_sole_of_foot	Human Skin	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	right sole of foot	FFO92CG	33	mimarks-survey	NA	FMA:Right foot surface	NA	NA	0	UBERON:zone of skin of foot	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F3	sample452	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F31Fotr	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	F3	right plantar foot	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFO92CG	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F31Fotr	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
F32Fotr.140582	GGACGTCACAGT	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	right_sole_of_foot	Human Skin	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	right sole of foot	FFO92CG	33	mimarks-survey	NA	FMA:Right foot surface	NA	NA	0	UBERON:zone of skin of foot	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F3	sample453	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F32Fotr	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	F3	right plantar foot	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFO92CG	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F32Fotr	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
F33Fotr.140629	GTATGTTGCTCA	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	right_sole_of_foot	Human Skin	source	NA	n	11/14/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	right sole of foot	FKB0RMH	33	mimarks-survey	NA	FMA:Right foot surface	NA	NA	0	UBERON:zone of skin of foot	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F3	sample454	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F33Fotr	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	F3	right plantar foot	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F33Fotr	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
F34Fotr.140673	GTCTCATGTAGG	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	right_sole_of_foot	Human Skin	source	NA	n	11/14/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	right sole of foot	FKB0RMH	33	mimarks-survey	NA	FMA:Right foot surface	NA	NA	0	UBERON:zone of skin of foot	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F3	sample455	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F34Fotr	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	F3	right plantar foot	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F34Fotr	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M11Fotr.140751	ACCAGCGACTAG	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	right_sole_of_foot	Human Skin	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	right sole of foot	FFLHOYS	36	mimarks-survey	NA	FMA:Right foot surface	NA	NA	0	UBERON:zone of skin of foot	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M1	sample456	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M11Fotr	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	36	M1	right plantar foot	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M11Fotr	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M12Fotr.140278	ACTGACAGCCAT	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	right_sole_of_foot	Human Skin	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	right sole of foot	FFLHOYS	36	mimarks-survey	NA	FMA:Right foot surface	NA	NA	0	UBERON:zone of skin of foot	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M1	sample457	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M12Fotr	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	36	M1	right plantar foot	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M12Fotr	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M13Fotr.140755	GAACTGTATCTC	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	right_sole_of_foot	Human Skin	source	NA	n	11/14/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	right sole of foot	FKB0RMH	36	mimarks-survey	NA	FMA:Right foot surface	NA	NA	0	UBERON:zone of skin of foot	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M1	sample458	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M13Fotr	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	36	M1	right plantar foot	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M13Fotr	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M14Fotr.140729	GACGAGTCAGTC	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	right_sole_of_foot	Human Skin	source	NA	n	11/14/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	right sole of foot	FKB0RMH	36	mimarks-survey	NA	FMA:Right foot surface	NA	NA	0	UBERON:zone of skin of foot	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M1	sample459	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M14Fotr	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	36	M1	right plantar foot	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M14Fotr	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M21Fotr.140859	CATGTAATGCTC	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	right_sole_of_foot	Human Skin	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	right sole of foot	FFO92CG	36	mimarks-survey	NA	FMA:Right foot surface	NA	NA	0	UBERON:zone of skin of foot	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M2	sample460	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M21Fotr	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	36	M2	right plantar foot	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFO92CG	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M21Fotr	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M22Fotr.140381	CGCAGACAGACT	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	right_sole_of_foot	Human Skin	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	right sole of foot	FFO92CG	36	mimarks-survey	NA	FMA:Right foot surface	NA	NA	0	UBERON:zone of skin of foot	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M2	sample461	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M22Fotr	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	36	M2	right plantar foot	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFO92CG	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M22Fotr	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M23Fotr.140344	GATCTCATAGGC	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	right_sole_of_foot	Human Skin	source	NA	n	11/14/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	right sole of foot	FKB0RMH	36	mimarks-survey	NA	FMA:Right foot surface	NA	NA	0	UBERON:zone of skin of foot	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M2	sample462	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M23Fotr	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	36	M2	right plantar foot	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M23Fotr	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M24Fotr.140398	GCAGCACGTTGA	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	right_sole_of_foot	Human Skin	source	NA	n	11/14/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	right sole of foot	FKB0RMH	36	mimarks-survey	NA	FMA:Right foot surface	NA	NA	0	UBERON:zone of skin of foot	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M2	sample463	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M24Fotr	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	36	M2	right plantar foot	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M24Fotr	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M31Fotr.140572	CTACACAAGCAC	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	right_sole_of_foot	Human Skin	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	right sole of foot	FFLHOYS	33	mimarks-survey	NA	FMA:Right foot surface	NA	NA	0	UBERON:zone of skin of foot	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M3	sample464	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M31Fotr	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	M3	right plantar foot	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M31Fotr	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M32Fotr.140562	CTGAGATACGCG	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	right_sole_of_foot	Human Skin	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	right sole of foot	FFLHOYS	33	mimarks-survey	NA	FMA:Right foot surface	NA	NA	0	UBERON:zone of skin of foot	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M3	sample465	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M32Fotr	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	M3	right plantar foot	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M32Fotr	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M33Fotr.140533	GGACGTCACAGT	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	right_sole_of_foot	Human Skin	source	NA	n	11/14/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	right sole of foot	FKB0RMH	33	mimarks-survey	NA	FMA:Right foot surface	NA	NA	0	UBERON:zone of skin of foot	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M3	sample466	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M33Fotr	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	M3	right plantar foot	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M33Fotr	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M34Fotr.140476	GTACTCTAGACT	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	right_sole_of_foot	Human Skin	source	NA	n	11/14/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	right sole of foot	FKB0RMH	33	mimarks-survey	NA	FMA:Right foot surface	NA	NA	0	UBERON:zone of skin of foot	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M3	sample467	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M34Fotr	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	M3	right plantar foot	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M34Fotr	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M41Fotr.140652	GACGAGTCAGTC	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	right_sole_of_foot	Human Skin	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	right sole of foot	FFO92CG	31	mimarks-survey	NA	FMA:Right foot surface	NA	NA	0	UBERON:zone of skin of foot	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M4	sample468	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M41Fotr	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	31	M4	right plantar foot	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFO92CG	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M41Fotr	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M42Fotr.140861	GATAGCTGTCTT	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	right_sole_of_foot	Human Skin	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	right sole of foot	FFO92CG	31	mimarks-survey	NA	FMA:Right foot surface	NA	NA	0	UBERON:zone of skin of foot	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M4	sample469	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M42Fotr	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	31	M4	right plantar foot	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FFO92CG	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M42Fotr	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M43Fotr.140475	GTGTACCTATCA	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	right_sole_of_foot	Human Skin	source	NA	n	11/14/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	right sole of foot	FKB0RMH	31	mimarks-survey	NA	FMA:Right foot surface	NA	NA	0	UBERON:zone of skin of foot	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M4	sample470	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M43Fotr	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	31	M4	right plantar foot	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M43Fotr	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M44Fotr.140394	TACACACATGGC	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	right_sole_of_foot	Human Skin	source	NA	n	11/14/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	right sole of foot	FKB0RMH	31	mimarks-survey	NA	FMA:Right foot surface	NA	NA	0	UBERON:zone of skin of foot	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M4	sample471	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M44Fotr	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	31	M4	right plantar foot	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M44Fotr	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M53Fotr.140451	TACTGCGACAGT	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	right_sole_of_foot	Human Skin	source	NA	n	11/14/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	right sole of foot	FKB0RMH	60	mimarks-survey	NA	FMA:Right foot surface	NA	NA	0	UBERON:zone of skin of foot	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M5	sample472	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M53Fotr	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	60	M5	right plantar foot	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M53Fotr	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M54Fotr.140667	TAGCGGATCACG	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	right_sole_of_foot	Human Skin	source	NA	n	11/14/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	right sole of foot	FKB0RMH	60	mimarks-survey	NA	FMA:Right foot surface	NA	NA	0	UBERON:zone of skin of foot	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M5	sample473	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M54Fotr	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	60	M5	right plantar foot	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M54Fotr	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M63Fotr.140467	TATGCACCAGTG	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	right_sole_of_foot	Human Skin	source	NA	n	11/14/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	right sole of foot	FKB0RMH	NA	mimarks-survey	NA	FMA:Right foot surface	NA	NA	0	UBERON:zone of skin of foot	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M6	sample474	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M63Fotr	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	NA	M6	right plantar foot	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M63Fotr	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
M64Fotr.140624	TCAGATCCGATG	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	right_sole_of_foot	Human Skin	source	NA	n	11/14/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	right sole of foot	FKB0RMH	NA	mimarks-survey	NA	FMA:Right foot surface	NA	NA	0	UBERON:zone of skin of foot	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M6	sample475	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:skin	UBERON:sebum	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M64Fotr	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	NA	M6	right plantar foot	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human skin metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	539655	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M64Fotr	ENVO:sebum	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human skin metagenome
F11Fcsw.140616	AGAGAGCAAGTG	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	stool	Human Feces	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	stool	FFLHOYS	33	mimarks-survey	NA	FMA:Feces	NA	NA	0	UBERON:feces	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F1	sample477	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:feces	UBERON:feces	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F11Fcsw	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	F1	stool	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human gut metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	408170	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F11Fcsw	ENVO:feces	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human gut metagenome
F12Fcsp.140694	AGTACTGCAGGC	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	stool	Human Feces	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	stool	FFLHOYS	33	mimarks-survey	NA	FMA:Feces	NA	NA	0	UBERON:feces	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F1	sample478	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:feces	UBERON:feces	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F12Fcsp	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	F1	stool	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human gut metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	408170	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F12Fcsp	ENVO:feces	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human gut metagenome
F12Fcsw.140400	AGTAGTATCCTC	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	stool	Human Feces	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	stool	FFLHOYS	33	mimarks-survey	NA	FMA:Feces	NA	NA	0	UBERON:feces	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F1	sample479	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:feces	UBERON:feces	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F12Fcsw	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	F1	stool	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human gut metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	408170	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F12Fcsw	ENVO:feces	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human gut metagenome
F13Fcsw.140387	GAGCAGATGCCT	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	stool	Human Feces	source	NA	n	11/14/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	stool	FKB0RMH	33	mimarks-survey	NA	FMA:Feces	NA	NA	0	UBERON:feces	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F1	sample480	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:feces	UBERON:feces	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F13Fcsw	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	F1	stool	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human gut metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	408170	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F13Fcsw	ENVO:feces	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human gut metagenome
F14Fcsw.140752	GATATGCGGCTG	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	stool	Human Feces	source	NA	n	11/14/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	stool	FKB0RMH	33	mimarks-survey	NA	FMA:Feces	NA	NA	0	UBERON:feces	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F1	sample481	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:feces	UBERON:feces	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F14Fcsw	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	F1	stool	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human gut metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	408170	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F14Fcsw	ENVO:feces	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human gut metagenome
F21Fcsp.140536	ATCCTCAGTAGT	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	stool	Human Feces	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	stool	FFLHOYS	31	mimarks-survey	NA	FMA:Feces	NA	NA	0	UBERON:feces	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F2	sample482	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:feces	UBERON:feces	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F21Fcsp	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	31	F2	stool	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human gut metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	408170	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F21Fcsp	ENVO:feces	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human gut metagenome
F21Fcsw.140506	ATCGATCTGTGG	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	stool	Human Feces	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	stool	FFLHOYS	31	mimarks-survey	NA	FMA:Feces	NA	NA	0	UBERON:feces	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F2	sample483	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:feces	UBERON:feces	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F21Fcsw	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	31	F2	stool	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human gut metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	408170	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F21Fcsw	ENVO:feces	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human gut metagenome
F22Fcsp.140783	CAACACGCACGA	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	stool	Human Feces	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	stool	FFLHOYS	31	mimarks-survey	NA	FMA:Feces	NA	NA	0	UBERON:feces	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F2	sample484	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:feces	UBERON:feces	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F22Fcsp	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	31	F2	stool	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human gut metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	408170	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F22Fcsp	ENVO:feces	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human gut metagenome
F22Fcsw.140281	CAACTATCAGCT	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	stool	Human Feces	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	stool	FFLHOYS	31	mimarks-survey	NA	FMA:Feces	NA	NA	0	UBERON:feces	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F2	sample485	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:feces	UBERON:feces	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F22Fcsw	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	31	F2	stool	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human gut metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	408170	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F22Fcsw	ENVO:feces	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human gut metagenome
F23Fcsw.140551	GCGTACAACTGT	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	stool	Human Feces	source	NA	n	11/14/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	stool	FKB0RMH	31	mimarks-survey	NA	FMA:Feces	NA	NA	0	UBERON:feces	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F2	sample486	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:feces	UBERON:feces	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F23Fcsw	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	31	F2	stool	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human gut metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	408170	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F23Fcsw	ENVO:feces	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human gut metagenome
F24Fcsw.140765	GCTGGTATCTGA	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	stool	Human Feces	source	NA	n	11/14/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	stool	FKB0RMH	31	mimarks-survey	NA	FMA:Feces	NA	NA	0	UBERON:feces	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F2	sample487	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:feces	UBERON:feces	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F24Fcsw	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	31	F2	stool	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human gut metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	408170	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F24Fcsw	ENVO:feces	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human gut metagenome
F31Fcsp.140571	GATCTTCAGTAC	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	stool	Human Feces	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	stool	FFO92CG	33	mimarks-survey	NA	FMA:Feces	NA	NA	0	UBERON:feces	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F3	sample488	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:feces	UBERON:feces	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F31Fcsp	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	F3	stool	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human gut metagenome	NA	ENVO:human-associated habitat	FLX	FFO92CG	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	408170	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F31Fcsp	ENVO:feces	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human gut metagenome
F31Fcsw.140492	GATGATCGCCGA	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	stool	Human Feces	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	stool	FFO92CG	33	mimarks-survey	NA	FMA:Feces	NA	NA	0	UBERON:feces	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F3	sample489	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:feces	UBERON:feces	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F31Fcsw	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	F3	stool	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human gut metagenome	NA	ENVO:human-associated habitat	FLX	FFO92CG	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	408170	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F31Fcsw	ENVO:feces	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human gut metagenome
F32Fcsp.140464	GCGGATGTGACT	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	stool	Human Feces	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	stool	FFO92CG	33	mimarks-survey	NA	FMA:Feces	NA	NA	0	UBERON:feces	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F3	sample490	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:feces	UBERON:feces	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F32Fcsp	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	F3	stool	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human gut metagenome	NA	ENVO:human-associated habitat	FLX	FFO92CG	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	408170	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F32Fcsp	ENVO:feces	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human gut metagenome
F32Fcsw.140706	GCGTACAACTGT	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	stool	Human Feces	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	stool	FFO92CG	33	mimarks-survey	NA	FMA:Feces	NA	NA	0	UBERON:feces	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F3	sample491	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:feces	UBERON:feces	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F32Fcsw	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	F3	stool	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human gut metagenome	NA	ENVO:human-associated habitat	FLX	FFO92CG	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	408170	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F32Fcsw	ENVO:feces	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human gut metagenome
F33Fcsw.140662	GTCGACTCCTCT	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	stool	Human Feces	source	NA	n	11/14/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	stool	FKB0RMH	33	mimarks-survey	NA	FMA:Feces	NA	NA	0	UBERON:feces	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F3	sample492	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:feces	UBERON:feces	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F33Fcsw	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	F3	stool	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human gut metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	408170	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F33Fcsw	ENVO:feces	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human gut metagenome
F34Fcsw.140733	GTGACTGCGGAT	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	stool	Human Feces	source	NA	n	11/14/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	stool	FKB0RMH	33	mimarks-survey	NA	FMA:Feces	NA	NA	0	UBERON:feces	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F3	sample493	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:feces	UBERON:feces	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F34Fcsw	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	F3	stool	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human gut metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	408170	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F34Fcsw	ENVO:feces	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human gut metagenome
M11Fcsp.140589	AATCGTGACTCG	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	stool	Human Feces	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	stool	FFLHOYS	36	mimarks-survey	NA	FMA:Feces	NA	NA	0	UBERON:feces	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M1	sample494	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:feces	UBERON:feces	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M11Fcsp	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	36	M1	stool	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human gut metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	408170	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M11Fcsp	ENVO:feces	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human gut metagenome
M11Fcsw.140610	ACACACTATGGC	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	stool	Human Feces	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	stool	FFLHOYS	36	mimarks-survey	NA	FMA:Feces	NA	NA	0	UBERON:feces	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M1	sample495	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:feces	UBERON:feces	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M11Fcsw	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	36	M1	stool	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human gut metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	408170	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M11Fcsw	ENVO:feces	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human gut metagenome
M12Fcsp.140668	ACGCGCAGATAC	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	stool	Human Feces	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	stool	FFLHOYS	36	mimarks-survey	NA	FMA:Feces	NA	NA	0	UBERON:feces	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M1	sample496	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:feces	UBERON:feces	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M12Fcsp	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	36	M1	stool	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human gut metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	408170	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M12Fcsp	ENVO:feces	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human gut metagenome
M12Fcsw.140326	ACGCTATCTGGA	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	stool	Human Feces	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	stool	FFLHOYS	36	mimarks-survey	NA	FMA:Feces	NA	NA	0	UBERON:feces	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M1	sample497	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:feces	UBERON:feces	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M12Fcsw	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	36	M1	stool	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human gut metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	408170	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M12Fcsw	ENVO:feces	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human gut metagenome
M13Fcsw.140390	GACAGCGTTGAC	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	stool	Human Feces	source	NA	n	11/14/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	stool	FKB0RMH	36	mimarks-survey	NA	FMA:Feces	NA	NA	0	UBERON:feces	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M1	sample498	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:feces	UBERON:feces	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M13Fcsw	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	36	M1	stool	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human gut metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	408170	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M13Fcsw	ENVO:feces	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human gut metagenome
M14Fcsw.140443	GACTAGACCAGC	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	stool	Human Feces	source	NA	n	11/14/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	stool	FKB0RMH	36	mimarks-survey	NA	FMA:Feces	NA	NA	0	UBERON:feces	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M1	sample499	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:feces	UBERON:feces	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M14Fcsw	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	36	M1	stool	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human gut metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	408170	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M14Fcsw	ENVO:feces	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human gut metagenome
M21Fcsp.140319	CAGGTGCTACTA	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	stool	Human Feces	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	stool	FFO92CG	36	mimarks-survey	NA	FMA:Feces	NA	NA	0	UBERON:feces	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M2	sample500	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:feces	UBERON:feces	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M21Fcsp	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	36	M2	stool	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human gut metagenome	NA	ENVO:human-associated habitat	FLX	FFO92CG	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	408170	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M21Fcsp	ENVO:feces	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human gut metagenome
M21Fcsw.140493	CAGTACGATCTT	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	stool	Human Feces	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	stool	FFO92CG	36	mimarks-survey	NA	FMA:Feces	NA	NA	0	UBERON:feces	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M2	sample501	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:feces	UBERON:feces	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M21Fcsw	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	36	M2	stool	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human gut metagenome	NA	ENVO:human-associated habitat	FLX	FFO92CG	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	408170	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M21Fcsw	ENVO:feces	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human gut metagenome
M22Fcsp.140527	CCTAGTACTGAT	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	stool	Human Feces	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	stool	FFO92CG	36	mimarks-survey	NA	FMA:Feces	NA	NA	0	UBERON:feces	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M2	sample502	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:feces	UBERON:feces	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M22Fcsp	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	36	M2	stool	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human gut metagenome	NA	ENVO:human-associated habitat	FLX	FFO92CG	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	408170	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M22Fcsp	ENVO:feces	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human gut metagenome
M22Fcsw.140276	CCTCTCGTGATC	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	stool	Human Feces	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	stool	FFO92CG	36	mimarks-survey	NA	FMA:Feces	NA	NA	0	UBERON:feces	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M2	sample503	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:feces	UBERON:feces	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M22Fcsw	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	36	M2	stool	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human gut metagenome	NA	ENVO:human-associated habitat	FLX	FFO92CG	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	408170	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M22Fcsw	ENVO:feces	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human gut metagenome
M23Fcsw.140821	GATTAGCACTCT	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	stool	Human Feces	source	NA	n	11/14/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	stool	FKB0RMH	36	mimarks-survey	NA	FMA:Feces	NA	NA	0	UBERON:feces	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M2	sample504	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:feces	UBERON:feces	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M23Fcsw	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	36	M2	stool	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human gut metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	408170	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M23Fcsw	ENVO:feces	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human gut metagenome
M24Fcsw.140647	GCATAGTAGCCG	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	stool	Human Feces	source	NA	n	11/14/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	stool	FKB0RMH	36	mimarks-survey	NA	FMA:Feces	NA	NA	0	UBERON:feces	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M2	sample505	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:feces	UBERON:feces	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M24Fcsw	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	36	M2	stool	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human gut metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	408170	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M24Fcsw	ENVO:feces	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human gut metagenome
M31Fcsp.140824	CGTAAGTCTACT	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	stool	Human Feces	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	stool	FFLHOYS	33	mimarks-survey	NA	FMA:Feces	NA	NA	0	UBERON:feces	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M3	sample506	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:feces	UBERON:feces	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M31Fcsp	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	M3	stool	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human gut metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	408170	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M31Fcsp	ENVO:feces	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human gut metagenome
M31Fcsw.140368	CGTACAGTTATC	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	stool	Human Feces	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	stool	FFLHOYS	33	mimarks-survey	NA	FMA:Feces	NA	NA	0	UBERON:feces	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M3	sample507	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:feces	UBERON:feces	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M31Fcsw	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	M3	stool	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human gut metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	408170	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M31Fcsw	ENVO:feces	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human gut metagenome
M32Fcsp.140470	CTAGTCAGCTGA	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	stool	Human Feces	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	stool	FFLHOYS	33	mimarks-survey	NA	FMA:Feces	NA	NA	0	UBERON:feces	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M3	sample508	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:feces	UBERON:feces	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M32Fcsp	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	M3	stool	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human gut metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	408170	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M32Fcsp	ENVO:feces	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human gut metagenome
M32Fcsw.140646	CTATAGTCGTGT	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	stool	Human Feces	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	stool	FFLHOYS	33	mimarks-survey	NA	FMA:Feces	NA	NA	0	UBERON:feces	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M3	sample509	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:feces	UBERON:feces	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M32Fcsw	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	M3	stool	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human gut metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	408170	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M32Fcsw	ENVO:feces	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human gut metagenome
M33Fcsw.140557	GGTATACGCAGC	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	stool	Human Feces	source	NA	n	11/14/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	stool	FKB0RMH	33	mimarks-survey	NA	FMA:Feces	NA	NA	0	UBERON:feces	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M3	sample510	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:feces	UBERON:feces	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M33Fcsw	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	M3	stool	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human gut metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	408170	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M33Fcsw	ENVO:feces	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human gut metagenome
M41Fcsp.140643	CTGTGACATTGT	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	stool	Human Feces	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	stool	FFO92CG	31	mimarks-survey	NA	FMA:Feces	NA	NA	0	UBERON:feces	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M4	sample511	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:feces	UBERON:feces	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M41Fcsp	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	31	M4	stool	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human gut metagenome	NA	ENVO:human-associated habitat	FLX	FFO92CG	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	408170	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M41Fcsp	ENVO:feces	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human gut metagenome
M41Fcsw.140359	CTGTGGATCGAT	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	stool	Human Feces	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	stool	FFO92CG	31	mimarks-survey	NA	FMA:Feces	NA	NA	0	UBERON:feces	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M4	sample512	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:feces	UBERON:feces	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M41Fcsw	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	31	M4	stool	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human gut metagenome	NA	ENVO:human-associated habitat	FLX	FFO92CG	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	408170	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M41Fcsw	ENVO:feces	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human gut metagenome
M42Fcsw.140447	GACTGCATCTTA	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	stool	Human Feces	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	stool	FFO92CG	31	mimarks-survey	NA	FMA:Feces	NA	NA	0	UBERON:feces	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M4	sample514	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:feces	UBERON:feces	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M42Fcsw	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	31	M4	stool	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human gut metagenome	NA	ENVO:human-associated habitat	FLX	FFO92CG	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	408170	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M42Fcsw	ENVO:feces	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human gut metagenome
M43Fcsw.140714	GTTCGCGTATAG	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	stool	Human Feces	source	NA	n	11/14/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	stool	FKB0RMH	31	mimarks-survey	NA	FMA:Feces	NA	NA	0	UBERON:feces	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M4	sample515	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:feces	UBERON:feces	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M43Fcsw	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	31	M4	stool	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human gut metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	408170	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M43Fcsw	ENVO:feces	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human gut metagenome
M44Fcsw.140418	TACGATGACCAC	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	stool	Human Feces	source	NA	n	11/14/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	stool	FKB0RMH	31	mimarks-survey	NA	FMA:Feces	NA	NA	0	UBERON:feces	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M4	sample516	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:feces	UBERON:feces	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M44Fcsw	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	31	M4	stool	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human gut metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	408170	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M44Fcsw	ENVO:feces	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human gut metagenome
M53Fcsw.140426	TAGATAGCAGGA	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	stool	Human Feces	source	NA	n	11/14/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	stool	FKB0RMH	60	mimarks-survey	NA	FMA:Feces	NA	NA	0	UBERON:feces	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M5	sample517	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:feces	UBERON:feces	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M53Fcsw	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	60	M5	stool	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human gut metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	408170	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M53Fcsw	ENVO:feces	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human gut metagenome
M54Fcsw.140330	TAGTGTGCTTCA	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	stool	Human Feces	source	NA	n	11/14/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	stool	FKB0RMH	60	mimarks-survey	NA	FMA:Feces	NA	NA	0	UBERON:feces	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M5	sample518	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:feces	UBERON:feces	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M54Fcsw	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	60	M5	stool	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human gut metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	408170	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M54Fcsw	ENVO:feces	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human gut metagenome
M63Fcsw.140818	TCACAGATCCGA	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	stool	Human Feces	source	NA	n	11/14/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	stool	FKB0RMH	NA	mimarks-survey	NA	FMA:Feces	NA	NA	0	UBERON:feces	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M6	sample519	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:feces	UBERON:feces	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M63Fcsw	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	NA	M6	stool	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human gut metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	408170	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M63Fcsw	ENVO:feces	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human gut metagenome
M64Fcsw.140360	TCAGTGACGTAC	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	stool	Human Feces	source	NA	n	11/14/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	stool	FKB0RMH	NA	mimarks-survey	NA	FMA:Feces	NA	NA	0	UBERON:feces	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M6	sample520	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:feces	UBERON:feces	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M64Fcsw	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	NA	M6	stool	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human gut metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	408170	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M64Fcsw	ENVO:feces	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human gut metagenome
F11Tong.140366	AGCTCCATACAG	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	tongue	Human Mouth	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	tongue	FFLHOYS	33	mimarks-survey	NA	FMA:Dorsal surface of tongue	NA	NA	0	UBERON:tongue	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F1	sample521	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:oral cavity	UBERON:saliva	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F11Tong	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	F1	dorsal surface of tongue	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human oral metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	447426	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F11Tong	ENVO:saliva	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human oral metagenome
F12Tong.140538	ATAGCTCCATAC	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	tongue	Human Mouth	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	tongue	FFLHOYS	33	mimarks-survey	NA	FMA:Dorsal surface of tongue	NA	NA	0	UBERON:tongue	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F1	sample522	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:oral cavity	UBERON:saliva	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F12Tong	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	F1	dorsal surface of tongue	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human oral metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	447426	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F12Tong	ENVO:saliva	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human oral metagenome
F13Tong.140677	GAGAGAATGATC	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	tongue	Human Mouth	source	NA	n	11/14/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	tongue	FKB0RMH	33	mimarks-survey	NA	FMA:Dorsal surface of tongue	NA	NA	0	UBERON:tongue	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F1	sample523	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:oral cavity	UBERON:saliva	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F13Tong	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	F1	dorsal surface of tongue	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human oral metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	447426	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F13Tong	ENVO:saliva	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human oral metagenome
F14Tong.140430	GATACGTCCTGA	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	tongue	Human Mouth	source	NA	n	11/14/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	tongue	FKB0RMH	33	mimarks-survey	NA	FMA:Dorsal surface of tongue	NA	NA	0	UBERON:tongue	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F1	sample524	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:oral cavity	UBERON:saliva	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F14Tong	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	F1	dorsal surface of tongue	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human oral metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	447426	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F14Tong	ENVO:saliva	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human oral metagenome
F21Tong.140655	ATGGCGTGCACA	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	tongue	Human Mouth	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	tongue	FFLHOYS	31	mimarks-survey	NA	FMA:Dorsal surface of tongue	NA	NA	0	UBERON:tongue	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F2	sample525	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:oral cavity	UBERON:saliva	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F21Tong	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	31	F2	dorsal surface of tongue	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human oral metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	447426	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F21Tong	ENVO:saliva	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human oral metagenome
F22Tong.140837	CAGACTCGCAGA	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	tongue	Human Mouth	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	tongue	FFLHOYS	31	mimarks-survey	NA	FMA:Dorsal surface of tongue	NA	NA	0	UBERON:tongue	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F2	sample526	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:oral cavity	UBERON:saliva	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F22Tong	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	31	F2	dorsal surface of tongue	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human oral metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	447426	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F22Tong	ENVO:saliva	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human oral metagenome
F23Tong.140417	GCGAGATCCAGT	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	tongue	Human Mouth	source	NA	n	11/14/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	tongue	FKB0RMH	31	mimarks-survey	NA	FMA:Dorsal surface of tongue	NA	NA	0	UBERON:tongue	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F2	sample527	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:oral cavity	UBERON:saliva	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F23Tong	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	31	F2	dorsal surface of tongue	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human oral metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	447426	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F23Tong	ENVO:saliva	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human oral metagenome
F24Tong.140829	GCTCGCTACTTC	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	tongue	Human Mouth	source	NA	n	11/14/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	tongue	FKB0RMH	31	mimarks-survey	NA	FMA:Dorsal surface of tongue	NA	NA	0	UBERON:tongue	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F2	sample528	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:oral cavity	UBERON:saliva	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F24Tong	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	31	F2	dorsal surface of tongue	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human oral metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	447426	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F24Tong	ENVO:saliva	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human oral metagenome
F31Tong.140785	GCATGTGCATGT	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	tongue	Human Mouth	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	tongue	FFO92CG	33	mimarks-survey	NA	FMA:Dorsal surface of tongue	NA	NA	0	UBERON:tongue	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F3	sample529	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:oral cavity	UBERON:saliva	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F31Tong	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	F3	dorsal surface of tongue	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human oral metagenome	NA	ENVO:human-associated habitat	FLX	FFO92CG	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	447426	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F31Tong	ENVO:saliva	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human oral metagenome
F32Tong.140410	GGATCGCAGATC	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	tongue	Human Mouth	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	tongue	FFO92CG	33	mimarks-survey	NA	FMA:Dorsal surface of tongue	NA	NA	0	UBERON:tongue	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F3	sample530	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:oral cavity	UBERON:saliva	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F32Tong	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	F3	dorsal surface of tongue	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human oral metagenome	NA	ENVO:human-associated habitat	FLX	FFO92CG	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	447426	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F32Tong	ENVO:saliva	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human oral metagenome
F33Tong.140358	GTCATATCGTAC	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	tongue	Human Mouth	source	NA	n	11/14/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	tongue	FKB0RMH	33	mimarks-survey	NA	FMA:Dorsal surface of tongue	NA	NA	0	UBERON:tongue	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F3	sample531	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:oral cavity	UBERON:saliva	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F33Tong	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	F3	dorsal surface of tongue	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human oral metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	447426	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F33Tong	ENVO:saliva	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human oral metagenome
F34Tong.140848	GTCTGGATAGCG	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	tongue	Human Mouth	source	NA	n	11/14/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	tongue	FKB0RMH	33	mimarks-survey	NA	FMA:Dorsal surface of tongue	NA	NA	0	UBERON:tongue	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F3	sample532	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:oral cavity	UBERON:saliva	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F34Tong	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	F3	dorsal surface of tongue	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human oral metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	447426	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F34Tong	ENVO:saliva	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human oral metagenome
M11Tong.140453	ACCGCAGAGTCA	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	tongue	Human Mouth	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	tongue	FFLHOYS	36	mimarks-survey	NA	FMA:Dorsal surface of tongue	NA	NA	0	UBERON:tongue	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M1	sample533	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:oral cavity	UBERON:saliva	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M11Tong	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	36	M1	dorsal surface of tongue	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human oral metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	447426	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M11Tong	ENVO:saliva	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human oral metagenome
M12Tong.140559	ACTGATCCTAGT	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	tongue	Human Mouth	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	tongue	FFLHOYS	36	mimarks-survey	NA	FMA:Dorsal surface of tongue	NA	NA	0	UBERON:tongue	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M1	sample534	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:oral cavity	UBERON:saliva	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M12Tong	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	36	M1	dorsal surface of tongue	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human oral metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	447426	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M12Tong	ENVO:saliva	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human oral metagenome
M13Tong.140491	GAAGTCTCGCAT	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	tongue	Human Mouth	source	NA	n	11/14/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	tongue	FKB0RMH	36	mimarks-survey	NA	FMA:Dorsal surface of tongue	NA	NA	0	UBERON:tongue	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M1	sample535	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:oral cavity	UBERON:saliva	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M13Tong	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	36	M1	dorsal surface of tongue	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human oral metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	447426	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M13Tong	ENVO:saliva	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human oral metagenome
M14Tong.140420	GACGCTAGTTCA	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	tongue	Human Mouth	source	NA	n	11/14/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	tongue	FKB0RMH	36	mimarks-survey	NA	FMA:Dorsal surface of tongue	NA	NA	0	UBERON:tongue	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M1	sample536	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:oral cavity	UBERON:saliva	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M14Tong	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	36	M1	dorsal surface of tongue	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human oral metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	447426	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M14Tong	ENVO:saliva	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human oral metagenome
M21Tong.140323	CATGTCTCTCCG	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	tongue	Human Mouth	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	tongue	FFO92CG	36	mimarks-survey	NA	FMA:Dorsal surface of tongue	NA	NA	0	UBERON:tongue	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M2	sample537	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:oral cavity	UBERON:saliva	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M21Tong	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	36	M2	dorsal surface of tongue	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human oral metagenome	NA	ENVO:human-associated habitat	FLX	FFO92CG	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	447426	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M21Tong	ENVO:saliva	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human oral metagenome
M22Tong.140364	CGCAGCGGTATA	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	tongue	Human Mouth	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	tongue	FFO92CG	36	mimarks-survey	NA	FMA:Dorsal surface of tongue	NA	NA	0	UBERON:tongue	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M2	sample538	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:oral cavity	UBERON:saliva	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M22Tong	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	36	M2	dorsal surface of tongue	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human oral metagenome	NA	ENVO:human-associated habitat	FLX	FFO92CG	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	447426	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M22Tong	ENVO:saliva	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human oral metagenome
M23Tong.140462	GATGCATGACGC	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	tongue	Human Mouth	source	NA	n	11/14/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	tongue	FKB0RMH	36	mimarks-survey	NA	FMA:Dorsal surface of tongue	NA	NA	0	UBERON:tongue	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M2	sample539	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:oral cavity	UBERON:saliva	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M23Tong	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	36	M2	dorsal surface of tongue	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human oral metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	447426	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M23Tong	ENVO:saliva	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human oral metagenome
M24Tong.140439	GCAGGCAGTACT	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	tongue	Human Mouth	source	NA	n	11/14/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	tongue	FKB0RMH	36	mimarks-survey	NA	FMA:Dorsal surface of tongue	NA	NA	0	UBERON:tongue	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M2	sample540	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:oral cavity	UBERON:saliva	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M24Tong	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	36	M2	dorsal surface of tongue	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human oral metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	447426	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M24Tong	ENVO:saliva	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human oral metagenome
M31Tong.140808	CTACATCTAAGC	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	tongue	Human Mouth	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	tongue	FFLHOYS	33	mimarks-survey	NA	FMA:Dorsal surface of tongue	NA	NA	0	UBERON:tongue	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M3	sample541	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:oral cavity	UBERON:saliva	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M31Tong	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	M3	dorsal surface of tongue	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human oral metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	447426	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M31Tong	ENVO:saliva	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human oral metagenome
M32Tong.140309	CTGAGCAGAGTC	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	tongue	Human Mouth	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	tongue	FFLHOYS	33	mimarks-survey	NA	FMA:Dorsal surface of tongue	NA	NA	0	UBERON:tongue	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M3	sample542	0.005	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:oral cavity	UBERON:saliva	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M32Tong	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	M3	dorsal surface of tongue	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human oral metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	447426	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M32Tong	ENVO:saliva	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human oral metagenome
M33Tong.140346	GGCGACATGTAC	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	tongue	Human Mouth	source	NA	n	11/14/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	tongue	FKB0RMH	33	mimarks-survey	NA	FMA:Dorsal surface of tongue	NA	NA	0	UBERON:tongue	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M3	sample543	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:oral cavity	UBERON:saliva	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M33Tong	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	M3	dorsal surface of tongue	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human oral metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	447426	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M33Tong	ENVO:saliva	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human oral metagenome
M34Tong.140540	GTAGATGCTTCG	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	tongue	Human Mouth	source	NA	n	11/14/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	tongue	FKB0RMH	33	mimarks-survey	NA	FMA:Dorsal surface of tongue	NA	NA	0	UBERON:tongue	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M3	sample544	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:oral cavity	UBERON:saliva	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M34Tong	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	M3	dorsal surface of tongue	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human oral metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	447426	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M34Tong	ENVO:saliva	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human oral metagenome
M41Tong.140317	GACGATATCGCG	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	tongue	Human Mouth	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	tongue	FFO92CG	31	mimarks-survey	NA	FMA:Dorsal surface of tongue	NA	NA	0	UBERON:tongue	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M4	sample545	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:oral cavity	UBERON:saliva	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M41Tong	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	31	M4	dorsal surface of tongue	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human oral metagenome	NA	ENVO:human-associated habitat	FLX	FFO92CG	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	447426	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M41Tong	ENVO:saliva	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human oral metagenome
M42Tong.140857	GATAGTGCCACT	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	tongue	Human Mouth	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	tongue	FFO92CG	31	mimarks-survey	NA	FMA:Dorsal surface of tongue	NA	NA	0	UBERON:tongue	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M4	sample546	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:oral cavity	UBERON:saliva	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M42Tong	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	31	M4	dorsal surface of tongue	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human oral metagenome	NA	ENVO:human-associated habitat	FLX	FFO92CG	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	447426	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M42Tong	ENVO:saliva	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human oral metagenome
M43Tong.140396	GTGTGTGTCAGG	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	tongue	Human Mouth	source	NA	n	11/14/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	tongue	FKB0RMH	31	mimarks-survey	NA	FMA:Dorsal surface of tongue	NA	NA	0	UBERON:tongue	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M4	sample547	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:oral cavity	UBERON:saliva	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M43Tong	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	31	M4	dorsal surface of tongue	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human oral metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	447426	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M43Tong	ENVO:saliva	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human oral metagenome
M44Tong.140642	TACAGTCTCATG	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	tongue	Human Mouth	source	NA	n	11/14/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	tongue	FKB0RMH	31	mimarks-survey	NA	FMA:Dorsal surface of tongue	NA	NA	0	UBERON:tongue	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M4	sample548	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:oral cavity	UBERON:saliva	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M44Tong	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	31	M4	dorsal surface of tongue	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human oral metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	447426	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M44Tong	ENVO:saliva	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human oral metagenome
M53Tong.140327	TACTTCGCTCGC	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	tongue	Human Mouth	source	NA	n	11/14/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	tongue	FKB0RMH	60	mimarks-survey	NA	FMA:Dorsal surface of tongue	NA	NA	0	UBERON:tongue	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M5	sample549	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:oral cavity	UBERON:saliva	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M53Tong	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	60	M5	dorsal surface of tongue	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human oral metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	447426	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M53Tong	ENVO:saliva	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human oral metagenome
M54Tong.140619	TAGGTATCTCAC	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	tongue	Human Mouth	source	NA	n	11/14/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	tongue	FKB0RMH	60	mimarks-survey	NA	FMA:Dorsal surface of tongue	NA	NA	0	UBERON:tongue	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M5	sample550	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:oral cavity	UBERON:saliva	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M54Tong	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	60	M5	dorsal surface of tongue	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human oral metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	447426	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M54Tong	ENVO:saliva	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human oral metagenome
M63Tong.140865	TATTCGTGTCAG	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	tongue	Human Mouth	source	NA	n	11/14/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	tongue	FKB0RMH	NA	mimarks-survey	NA	FMA:Dorsal surface of tongue	NA	NA	0	UBERON:tongue	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M6	sample551	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:oral cavity	UBERON:saliva	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M63Tong	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	NA	M6	dorsal surface of tongue	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human oral metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	447426	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M63Tong	ENVO:saliva	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human oral metagenome
M64Tong.140427	TCAGGACTGTGT	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	tongue	Human Mouth	source	NA	n	11/14/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	tongue	FKB0RMH	NA	mimarks-survey	NA	FMA:Dorsal surface of tongue	NA	NA	0	UBERON:tongue	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M6	sample552	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:oral cavity	UBERON:saliva	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M64Tong	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	NA	M6	dorsal surface of tongue	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human oral metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial Community Variation in Human Body Habitats Across Space and Time	447426	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M64Tong	ENVO:saliva	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human oral metagenome
F11Urin.140632	AGCTCTCAGAGG	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	urine	Human Urine	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	urine	FFLHOYS	33	mimarks-survey	NA	FMA:Urine	NA	NA	0	UBERON:urine	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F1	sample553	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:urine	UBERON:urine	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F11Urin	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	F1	urine	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human urine metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial variation in human body habitats across space and time	646099	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F11Urin	ENVO:urine	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human urine metagenome
F12Urin.140382	ATAGGCGATCTC	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	urine	Human Urine	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	urine	FFLHOYS	33	mimarks-survey	NA	FMA:Urine	NA	NA	0	UBERON:urine	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F1	sample554	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:urine	UBERON:urine	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F12Urin	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	F1	urine	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human urine metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial variation in human body habitats across space and time	646099	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F12Urin	ENVO:urine	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human urine metagenome
F13Uric.140806	GAGATGCCGACT	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	urine	Human Urine	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	urine	FKB0RMH	33	mimarks-survey	NA	FMA:Urine	NA	NA	0	UBERON:urine	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F1	sample567	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:urine	UBERON:urine	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F13Uric	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	F1	urine	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human urine metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial variation in human body habitats across space and time	646099	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F13Uric	ENVO:urine	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human urine metagenome
F13Urin.140615	GAGAGCTCTACG	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	urine	Human Urine	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	urine	FKB0RMH	33	mimarks-survey	NA	FMA:Urine	NA	NA	0	UBERON:urine	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F1	sample568	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:urine	UBERON:urine	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F13Urin	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	F1	urine	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human urine metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial variation in human body habitats across space and time	646099	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F13Urin	ENVO:urine	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human urine metagenome
F14Uric.140596	GATAGTGCCACT	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	urine	Human Urine	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	urine	FKB0RMH	33	mimarks-survey	NA	FMA:Urine	NA	NA	0	UBERON:urine	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F1	sample569	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:urine	UBERON:urine	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F14Uric	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	F1	urine	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human urine metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial variation in human body habitats across space and time	646099	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F14Uric	ENVO:urine	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human urine metagenome
F14Urin.140297	GATAGCTGTCTT	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	urine	Human Urine	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	urine	FKB0RMH	33	mimarks-survey	NA	FMA:Urine	NA	NA	0	UBERON:urine	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F1	sample570	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:urine	UBERON:urine	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F14Urin	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	F1	urine	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human urine metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial variation in human body habitats across space and time	646099	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F14Urin	ENVO:urine	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human urine metagenome
F21Urin.140569	ATGGTCTACTAC	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	urine	Human Urine	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	urine	FFLHOYS	31	mimarks-survey	NA	FMA:Urine	NA	NA	0	UBERON:urine	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F2	sample555	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:urine	UBERON:urine	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F21Urin	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	31	F2	urine	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human urine metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial variation in human body habitats across space and time	646099	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F21Urin	ENVO:urine	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human urine metagenome
F22Urin.140306	CAGAGGAGCTCT	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	urine	Human Urine	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	urine	FFLHOYS	31	mimarks-survey	NA	FMA:Urine	NA	NA	0	UBERON:urine	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F2	sample556	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:urine	UBERON:urine	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F22Urin	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	31	F2	urine	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human urine metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial variation in human body habitats across space and time	646099	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F22Urin	ENVO:urine	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human urine metagenome
F23Uric.140274	GCGGATGTGACT	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	urine	Human Urine	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	urine	FKB0RMH	31	mimarks-survey	NA	FMA:Urine	NA	NA	0	UBERON:urine	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F2	sample571	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:urine	UBERON:urine	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F23Uric	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	31	F2	urine	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human urine metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial variation in human body habitats across space and time	646099	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F23Uric	ENVO:urine	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human urine metagenome
F23Urin.140495	GCGATATATCGC	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	urine	Human Urine	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	urine	FKB0RMH	31	mimarks-survey	NA	FMA:Urine	NA	NA	0	UBERON:urine	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F2	sample572	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:urine	UBERON:urine	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F23Urin	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	31	F2	urine	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human urine metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial variation in human body habitats across space and time	646099	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F23Urin	ENVO:urine	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human urine metagenome
F24Uric.140658	GCTGCTGCAATA	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	urine	Human Urine	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	urine	FKB0RMH	31	mimarks-survey	NA	FMA:Urine	NA	NA	0	UBERON:urine	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F2	sample573	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:urine	UBERON:urine	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F24Uric	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	31	F2	urine	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human urine metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial variation in human body habitats across space and time	646099	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F24Uric	ENVO:urine	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human urine metagenome
F24Urin.140403	GCTGATGAGCTG	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	urine	Human Urine	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	urine	FKB0RMH	31	mimarks-survey	NA	FMA:Urine	NA	NA	0	UBERON:urine	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F2	sample574	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:urine	UBERON:urine	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F24Urin	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	31	F2	urine	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human urine metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial variation in human body habitats across space and time	646099	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F24Urin	ENVO:urine	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human urine metagenome
F31Urin.140602	GCATTGCGTGAG	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	urine	Human Urine	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	urine	FFO92CG	33	mimarks-survey	NA	FMA:Urine	NA	NA	0	UBERON:urine	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F3	sample561	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:urine	UBERON:urine	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F31Urin	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	F3	urine	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human urine metagenome	NA	ENVO:human-associated habitat	FLX	FFO92CG	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial variation in human body habitats across space and time	646099	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F31Urin	ENVO:urine	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human urine metagenome
F32Urin.140725	GGCAGTGTATCG	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	urine	Human Urine	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	urine	FFO92CG	33	mimarks-survey	NA	FMA:Urine	NA	NA	0	UBERON:urine	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F3	sample562	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:urine	UBERON:urine	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F32Urin	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	F3	urine	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human urine metagenome	NA	ENVO:human-associated habitat	FLX	FFO92CG	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial variation in human body habitats across space and time	646099	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F32Urin	ENVO:urine	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human urine metagenome
F33Uric.140793	GTCCATAGCTAG	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	urine	Human Urine	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	urine	FKB0RMH	33	mimarks-survey	NA	FMA:Urine	NA	NA	0	UBERON:urine	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F3	sample575	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:urine	UBERON:urine	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F33Uric	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	F3	urine	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human urine metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial variation in human body habitats across space and time	646099	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F33Uric	ENVO:urine	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human urine metagenome
F33Urin.140518	GTCATTCACGAG	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	urine	Human Urine	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	urine	FKB0RMH	33	mimarks-survey	NA	FMA:Urine	NA	NA	0	UBERON:urine	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F3	sample576	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:urine	UBERON:urine	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F33Urin	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	F3	urine	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human urine metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial variation in human body habitats across space and time	646099	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F33Urin	ENVO:urine	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human urine metagenome
F34Uric.140423	GTGACCTGATGT	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	urine	Human Urine	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	urine	FKB0RMH	33	mimarks-survey	NA	FMA:Urine	NA	NA	0	UBERON:urine	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F3	sample577	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:urine	UBERON:urine	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F34Uric	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	F3	urine	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human urine metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial variation in human body habitats across space and time	646099	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F34Uric	ENVO:urine	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human urine metagenome
F34Urin.140432	GTCTTCGTCGCT	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	urine	Human Urine	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	urine	FKB0RMH	33	mimarks-survey	NA	FMA:Urine	NA	NA	0	UBERON:urine	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	female	NA	NA	NA	NA	449:F3	sample578	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:urine	UBERON:urine	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	F34Urin	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	F3	urine	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human urine metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial variation in human body habitats across space and time	646099	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	F34Urin	ENVO:urine	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human urine metagenome
M11Urin.140484	ACCTCGATCAGA	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	urine	Human Urine	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	urine	FFLHOYS	36	mimarks-survey	NA	FMA:Urine	NA	NA	0	UBERON:urine	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M1	sample557	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:urine	UBERON:urine	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M11Urin	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	36	M1	urine	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human urine metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial variation in human body habitats across space and time	646099	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M11Urin	ENVO:urine	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human urine metagenome
M13Uric.140340	GACACTCGAATC	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	urine	Human Urine	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	urine	FKB0RMH	36	mimarks-survey	NA	FMA:Urine	NA	NA	0	UBERON:urine	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M1	sample579	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:urine	UBERON:urine	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M13Uric	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	36	M1	urine	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human urine metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial variation in human body habitats across space and time	646099	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M13Uric	ENVO:urine	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human urine metagenome
M13Urin.140575	GAATGATGAGTG	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	urine	Human Urine	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	urine	FKB0RMH	36	mimarks-survey	NA	FMA:Urine	NA	NA	0	UBERON:urine	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M1	sample580	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:urine	UBERON:urine	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M13Urin	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	36	M1	urine	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human urine metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial variation in human body habitats across space and time	646099	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M13Urin	ENVO:urine	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human urine metagenome
M14Uric.140532	GACTAACGTCAC	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	urine	Human Urine	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	urine	FKB0RMH	36	mimarks-survey	NA	FMA:Urine	NA	NA	0	UBERON:urine	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M1	sample581	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:urine	UBERON:urine	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M14Uric	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	36	M1	urine	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human urine metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial variation in human body habitats across space and time	646099	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M14Uric	ENVO:urine	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human urine metagenome
M14Urin.140599	GACGTTGCACAG	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	urine	Human Urine	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	urine	FKB0RMH	36	mimarks-survey	NA	FMA:Urine	NA	NA	0	UBERON:urine	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M1	sample582	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:urine	UBERON:urine	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M14Urin	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	36	M1	urine	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human urine metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial variation in human body habitats across space and time	646099	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M14Urin	ENVO:urine	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human urine metagenome
M21Urin.140335	CATTCGATGACT	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	urine	Human Urine	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	urine	FFO92CG	36	mimarks-survey	NA	FMA:Urine	NA	NA	0	UBERON:urine	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M2	sample563	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:urine	UBERON:urine	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M21Urin	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	36	M2	urine	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human urine metagenome	NA	ENVO:human-associated habitat	FLX	FFO92CG	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial variation in human body habitats across space and time	646099	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M21Urin	ENVO:urine	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human urine metagenome
M22Urin.140805	CGCATGAGGATC	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	urine	Human Urine	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	urine	FFO92CG	36	mimarks-survey	NA	FMA:Urine	NA	NA	0	UBERON:urine	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M2	sample564	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:urine	UBERON:urine	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M22Urin	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	36	M2	urine	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human urine metagenome	NA	ENVO:human-associated habitat	FLX	FFO92CG	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial variation in human body habitats across space and time	646099	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M22Urin	ENVO:urine	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human urine metagenome
M23Uric.140523	GATGTGAGCGCT	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	urine	Human Urine	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	urine	FKB0RMH	36	mimarks-survey	NA	FMA:Urine	NA	NA	0	UBERON:urine	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M2	sample583	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:urine	UBERON:urine	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M23Uric	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	36	M2	urine	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human urine metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial variation in human body habitats across space and time	646099	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M23Uric	ENVO:urine	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human urine metagenome
M23Urin.140740	GATGTCGTGTCA	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	urine	Human Urine	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	urine	FKB0RMH	36	mimarks-survey	NA	FMA:Urine	NA	NA	0	UBERON:urine	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M2	sample584	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:urine	UBERON:urine	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M23Urin	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	36	M2	urine	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human urine metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial variation in human body habitats across space and time	646099	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M23Urin	ENVO:urine	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human urine metagenome
M24Uric.140635	GCAGTTCATATC	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	urine	Human Urine	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	urine	FKB0RMH	36	mimarks-survey	NA	FMA:Urine	NA	NA	0	UBERON:urine	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M2	sample585	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:urine	UBERON:urine	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M24Uric	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	36	M2	urine	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human urine metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial variation in human body habitats across space and time	646099	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M24Uric	ENVO:urine	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human urine metagenome
M24Urin.140388	GCAGTATCACTG	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	urine	Human Urine	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	urine	FKB0RMH	36	mimarks-survey	NA	FMA:Urine	NA	NA	0	UBERON:urine	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M2	sample586	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:urine	UBERON:urine	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M24Urin	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	36	M2	urine	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human urine metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial variation in human body habitats across space and time	646099	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M24Urin	ENVO:urine	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human urine metagenome
M31Urin.140563	CTACGCGTCTCT	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	urine	Human Urine	source	NA	n	8/18/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	urine	FFLHOYS	33	mimarks-survey	NA	FMA:Urine	NA	NA	0	UBERON:urine	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M3	sample559	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:urine	UBERON:urine	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M31Urin	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	M3	urine	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human urine metagenome	NA	ENVO:human-associated habitat	FLX	FFLHOYS	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial variation in human body habitats across space and time	646099	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M31Urin	ENVO:urine	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human urine metagenome
M33Uric.140581	GGCTATGACATC	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	urine	Human Urine	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	urine	FKB0RMH	33	mimarks-survey	NA	FMA:Urine	NA	NA	0	UBERON:urine	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M3	sample587	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:urine	UBERON:urine	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M33Uric	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	M3	urine	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human urine metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial variation in human body habitats across space and time	646099	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M33Uric	ENVO:urine	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human urine metagenome
M33Urin.140845	GGCGTACTGATG	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	urine	Human Urine	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	urine	FFO92CG	33	mimarks-survey	NA	FMA:Urine	NA	NA	0	UBERON:urine	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M3	sample588	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:urine	UBERON:urine	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M33Urin	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	M3	urine	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human urine metagenome	NA	ENVO:human-associated habitat	FLX	FFO92CG	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial variation in human body habitats across space and time	646099	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M33Urin	ENVO:urine	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human urine metagenome
M34Urin.140663	GTAGCAACGTCT	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	urine	Human Urine	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	urine	FKB0RMH	33	mimarks-survey	NA	FMA:Urine	NA	NA	0	UBERON:urine	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M3	sample590	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:urine	UBERON:urine	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M34Urin	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	33	M3	urine	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human urine metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial variation in human body habitats across space and time	646099	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M34Urin	ENVO:urine	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human urine metagenome
M41Urin.140696	GACGCAGTAGCT	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	urine	Human Urine	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	urine	FFO92CG	31	mimarks-survey	NA	FMA:Urine	NA	NA	0	UBERON:urine	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M4	sample565	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:urine	UBERON:urine	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M41Urin	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	31	M4	urine	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human urine metagenome	NA	ENVO:human-associated habitat	FLX	FFO92CG	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial variation in human body habitats across space and time	646099	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M41Urin	ENVO:urine	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human urine metagenome
M42Urin.140778	GATATGCGGCTG	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	urine	Human Urine	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	urine	FFO92CG	31	mimarks-survey	NA	FMA:Urine	NA	NA	0	UBERON:urine	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M4	sample566	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:urine	UBERON:urine	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M42Urin	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	31	M4	urine	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human urine metagenome	NA	ENVO:human-associated habitat	FLX	FFO92CG	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial variation in human body habitats across space and time	646099	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M42Urin	ENVO:urine	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human urine metagenome
M43Urin.140463	GTGTTGCAGCAT	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	urine	Human Urine	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	urine	FKB0RMH	31	mimarks-survey	NA	FMA:Urine	NA	NA	0	UBERON:urine	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M4	sample592	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:urine	UBERON:urine	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M43Urin	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	31	M4	urine	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human urine metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial variation in human body habitats across space and time	646099	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M43Urin	ENVO:urine	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human urine metagenome
M44Uric.140688	TACCGCTAGTAG	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	urine	Human Urine	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	urine	FKB0RMH	31	mimarks-survey	NA	FMA:Urine	NA	NA	0	UBERON:urine	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M4	sample593	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:urine	UBERON:urine	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M44Uric	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	31	M4	urine	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human urine metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial variation in human body habitats across space and time	646099	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M44Uric	ENVO:urine	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human urine metagenome
M44Urin.140414	TACATCACCACA	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	urine	Human Urine	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	urine	FKB0RMH	31	mimarks-survey	NA	FMA:Urine	NA	NA	0	UBERON:urine	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M4	sample594	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:urine	UBERON:urine	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M44Urin	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	31	M4	urine	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human urine metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial variation in human body habitats across space and time	646099	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M44Urin	ENVO:urine	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human urine metagenome
M53Uric.140515	TAGAGAGAGTGG	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	urine	Human Urine	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	urine	FKB0RMH	60	mimarks-survey	NA	FMA:Urine	NA	NA	0	UBERON:urine	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M5	sample595	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:urine	UBERON:urine	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M53Uric	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	60	M5	urine	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human urine metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial variation in human body habitats across space and time	646099	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M53Uric	ENVO:urine	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human urine metagenome
M53Urin.140277	TAGACTGTACTC	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	urine	Human Urine	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	urine	FKB0RMH	60	mimarks-survey	NA	FMA:Urine	NA	NA	0	UBERON:urine	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M5	sample596	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:urine	UBERON:urine	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M53Urin	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	60	M5	urine	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human urine metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial variation in human body habitats across space and time	646099	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M53Urin	ENVO:urine	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human urine metagenome
M54Uric.140775	TAGTGCTGCGTA	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	urine	Human Urine	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	urine	FKB0RMH	60	mimarks-survey	NA	FMA:Urine	NA	NA	0	UBERON:urine	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M5	sample597	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:urine	UBERON:urine	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M54Uric	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	60	M5	urine	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human urine metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial variation in human body habitats across space and time	646099	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M54Uric	ENVO:urine	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human urine metagenome
M54Urin.140480	TAGTCGTCTAGT	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	urine	Human Urine	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	urine	FKB0RMH	60	mimarks-survey	NA	FMA:Urine	NA	NA	0	UBERON:urine	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M5	sample598	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:urine	UBERON:urine	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M54Urin	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	60	M5	urine	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human urine metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial variation in human body habitats across space and time	646099	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M54Urin	ENVO:urine	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human urine metagenome
M63Uric.140617	TCAATCTAGCGT	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	urine	Human Urine	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	urine	FKB0RMH	NA	mimarks-survey	NA	FMA:Urine	NA	NA	0	UBERON:urine	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M6	sample599	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:urine	UBERON:urine	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M63Uric	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	NA	M6	urine	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human urine metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial variation in human body habitats across space and time	646099	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M63Uric	ENVO:urine	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human urine metagenome
M63Urin.140645	TCAACAGCATCG	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	urine	Human Urine	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	urine	FKB0RMH	NA	mimarks-survey	NA	FMA:Urine	NA	NA	0	UBERON:urine	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M6	sample600	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:urine	UBERON:urine	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M63Urin	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	NA	M6	urine	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human urine metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial variation in human body habitats across space and time	646099	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M63Urin	ENVO:urine	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human urine metagenome
M64Uric.140672	TCAGTCGACGAG	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	urine	Human Urine	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	urine	FKB0RMH	NA	mimarks-survey	NA	FMA:Urine	NA	NA	0	UBERON:urine	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M6	sample601	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:urine	UBERON:urine	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M64Uric	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	NA	M6	urine	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human urine metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial variation in human body habitats across space and time	646099	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M64Uric	ENVO:urine	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human urine metagenome
M64Urin.140764	TCAGTACGAGGC	CATGCTGCCTCCCGTAGGAGT	costelle@stanford.edu	swab with sterile saline	NA	urine	Human Urine	source	NA	n	8/20/08	Bacterial Community Variation in Human Body Habitats Across Space and Time	urine	FKB0RMH	NA	mimarks-survey	NA	FMA:Urine	NA	NA	0	UBERON:urine	NA	1624.097656	2008-2009	NA	0	PMID: 19892944	initial denaturation: 94; annealing: 50_0.5; elongation: 72_1.5; final elongation: 72_10; 35	male	NA	NA	NA	NA	449:M6	sample602	0.006	"1, swab"	NA	http://www.scienceonline.org/cgi/content/abstract/1177486	Sampling of multiple human body sites	NA	NA	-105.2705456	y	NA	CCME	http://www.mobio.com/soil-dna-isolation/powersoil-dna-isolation-kit.html	UBERON:urine	UBERON:urine	"Elucidating the biogeography of bacterial communities on the human body is critical for establishing healthy baselines from which to detect differences associated with diseases. To obtain an integrated view of the spatial and temporal distribution of the human microbiota, we surveyed bacteria from up to 27 sites in seven to nine healthy adults on four occasions. We found that community composition was determined primarily by body habitat. Within habitats, interpersonal variability was high, whereas individuals exhibited minimal temporal variability. Several skin locations harbored more diverse communities than the gut and mouth, and skin locations differed in their community assembly patterns. These results indicate that our microbiota, although personalized, varies systematically across body habitats and time; such trends may ultimately reveal how microbiome changes cause or prevent disease."	ENVO:human-associated habitat	TCAG	Bacterial community variation in human body habitats across space and time	M64Urin	0	"Healthy adults were recruited to donate samples on two consecutive days in June and again three months later on two consecutive days in September 2008. Volunteers were unrelated individuals of both sexes and most lived in Boulder, CO. Female subjects F1, F2, F3 and male subjects M1, M2, M3, M4, and M6 were 30-35 years old. Subject M5 was approximately 60 years old. The health status of the volunteers was self-reported. None of the volunteers indicated that they had been hospitalized in the previous 6 months. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0708.12). On the day of sampling, participants were instructed to bathe in the morning using their choice of personal hygiene products, but to avoid wearing deodorant or antiperspirant. The interval between subject bathing and the collection of most samples likely ranged from around several hours to half an hour. Participants were asked to avoid eating or drinking for one hour prior to oral sampling. Skin surfaces, hair on top of the head, and nostrils (nares) were sampled using autoclaved cotton-tipped swabs (Puritan) pre-moistened with a sterile solution of 0.15 M NaCl and 0.1% Tween 20 (S1, S2). Swabbing has been shown to be a suitable method for skin sample collection for microbial community analysis (S3). Dry cotton-tipped swabs were used to sample the external auditory canal [and earwax (cerumen) if present], the dorsal surface of the tongue, and stool. Stool samples have been shown to provide an adequately representative sample of the gut microbiota (S4). Stool was swabbed using two different methods for comparison: i) off of used bathroom tissue and ii) by stabbing the swab directly into the largest piece of feces (see Fig. S22 for comparison). We found that swabbing off of used bathroom tissue was adequate for obtaining material for microbial community analysis. Swabs were placed in sterile 15 ml plastic tubes and stored at -80C within 10 minutes of sampling. Volunteers sampled stool on the same days that the other samples were collected (within 24 hours). If needed, the subjects were provided dry ice for the transport of their stool samples back to the lab. Liquid samples were also collected. The oral cavity was rinsed by swishing 10 ml of sterile water for 10 seconds, and then spitting the water into a sterile 50 ml plastic tube (see Fig. S22 for comparison to dorsal tongue swabs). Liquids were stored at 4C for several hours before sterile filtration onto 0.2micro-m pore size cellulose-nitrate filters (Nalgene). Filtered samples were immediately placed at -80C. Volunteers sampled themselves while wearing sterile gloves after first having their palms and index fingers swabbed by another person wearing sterile gloves. Replicate samples from the right and left sides of the body were taken when possible. All but one subject had not taken antibiotics in the 6 months prior to the June sampling dates, and none took antibiotics during the study period. Figure S1 lists the sample types, dates, and subjects. Samples were stored at -80C for <1 week prior to DNA extraction. We performed a follow-up study aimed at addressing whether skin location influences bacterial community assembly. Our approach involved transplanting foreign bacteria from various donor sites (dorsal tongue, forehead, and volar forearm) onto disinfected recipient skin sites (forehead and volar forearm) both within and between individuals. All individuals were made aware of the nature of the experiment and gave written informed consent in accordance with the sampling protocol approved by the University of Colorado Human Research Committee (protocol 0109.23)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19892944	NA	M6	urine	Elizabeth Costello	CCME	GAZ:United States of America	NA	CostelloWholeBodySites	0	NA	9606	NA	NA	n	NA	NA	Costello_whole_body_sites	pyrosequencing	NA	NA	NA	human urine metagenome	NA	ENVO:human-associated habitat	FLX	FKB0RMH	NA	NA	NA	NA	CostelloWholeBodySites	Bacterial variation in human body habitats across space and time	646099	CCME	NA	years	NA	449	"Amplicon DNA concentrations were determined using the Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen). Assays were carried out using 5-10 _l of cleaned PCR product in a total reaction volume of 200 _l in black, 96-well microtiter plates. Fluorescence was measured on a BioTek Synergy HT plate reader using the 480/520-nm excitation/emission filter pair. Following quantitation, cleaned amplicons were combined in equimolar ratios into a single tube. The final pool of DNA was precipitated on ice for 45 minutes following theaddition of 5 M NaCl (0.2 M final concentration) and 2 volumes of ice-cold 100% ethanol. The precipitated DNA was centrifuged at 7,800 x g for 40 minutes at 4C, and the resulting pellet was washed with an equal volume of ice-cold 70% ethanol and centrifuged again at 7,800 x g for 20 minutes at 4C. The supernatant was removed and the pellet was air dried for 10 minutes at room temperature and then resuspended in 100 micro-l of nuclease-free water (MO BIO). The final concentration of the pooled DNA was determined using a NanoDrop spectrophotometer (Thermo Fisher). Pyrosequencing was carried out using primer A on a 454 Life Sciences Genome Sequencer FLX instrument (Roche)."	costello_whole_body_sites	NA	y	human	M64Urin	ENVO:urine	16S rRNA	NA	NA	V2	CGS-GL	V2	NA	NA	NULL	NA	40.0149856	NA	NA	human urine metagenome
MBD10.660775	CGTTATGTACAC	CATGCTGCCTCCCGTAGGAGT	NA	NA	NICU 1	BabyBedside	NICU BabyBedside	sink	NA	n	5/28/09	Kelley_nicu_contamination	NA	FWIRNKE02	NA	mimarks-survey	NA	NA	BabyBedside	NA	0	NA	NA	246.888	2/1/09	NA	0	NA	NA	NA	NA	NA	NA	drscottkelley@gmail.com	NA	NA	0.025	1	NA	NA	This study examined contamination of inanimate hospital environments as possible sources of contamination in Newborn Intensive Care Units	NA	NA	-117.157255	y	NA	SDSU	NA	NA	NA	This study examined contamination of inanimate hospital environments as possible sources of contamination in Newborn Intensive Care Units	ENVO:anthropogenic habitat	TCAG	Evidence of widespread opportunistic pathogen contamination in Newborn Intensive Care Units	FWIRNKE02_CGTTATGTACAC	2	Surface contamination swab sample from Newborn Intensive Care Unit assessing bacterial diversity on inanimate hospital environments.	FWD:AGAGTTTGATCCTGGCTCAG; REV:TGCTGCCTCCCGTAGGAGT	NA	NA	NA	NA	Dan Knights/Antonio	SDSU	GAZ:United States of America	NA	contamination_nicu_FWIRNKE02	0	NA	NA	NA	NA	n	NICU 1	Scott Kelley	Kelley_Newborn_ICU_Study	pyrosequencing	NA	NA	NA	unclassified metagenome	NA	ENVO:anthropogenic habitat	FLX	contamination_nicu_FWIRNKE02	NA	NA	NA	NA	nicu_contamination_data	Contamination in Newborn Intensive Care Units	408169	CCME	NA	NA	NA	386	"DNA was extracted using a lysozyme extraction followed by extraction with a QIAGEN DNEasy Blood and Tissue Extraction Kit. Each amplicon library was constructed by amplifying the16S rRNA gene using the modified 27F and and the 338R primer pair. Primers contained DNA barcode sequences such as those described by Fierer et al 2008, and the recommended 454 adapter sequences. Amplification conditions were exactly as described Fierer et al 2008."	contamination_nicu	NA	n	NA	FWIRNKE02_MBD10	ENVO:surface	16S rRNA	NA	NA	V2	CCME	V1-V2	DiaperCounter	NA	NA	NA	32.715329	NA	NICU1	Swab_was_taken_from_NICU_baby_open_bedside_diaper_counter_machine_NICU1
MBP17.660767	CTACTGATATCG	CATGCTGCCTCCCGTAGGAGT	NA	NA	NICU 1	BabyBedside	NICU BabyBedside	sink	NA	n	5/28/09	Kelley_nicu_contamination	NA	FWIRNKE02	NA	mimarks-survey	NA	NA	BabyBedside	NA	0	NA	NA	246.888	2/1/09	NA	0	NA	NA	NA	NA	NA	NA	drscottkelley@gmail.com	NA	NA	0.025	1	NA	NA	This study examined contamination of inanimate hospital environments as possible sources of contamination in Newborn Intensive Care Units	NA	NA	-117.157255	y	NA	SDSU	NA	NA	NA	This study examined contamination of inanimate hospital environments as possible sources of contamination in Newborn Intensive Care Units	ENVO:anthropogenic habitat	TCAG	Evidence of widespread opportunistic pathogen contamination in Newborn Intensive Care Units	FWIRNKE02_CTACTGATATCG	2	Surface contamination swab sample from Newborn Intensive Care Unit assessing bacterial diversity on inanimate hospital environments.	FWD:AGAGTTTGATCCTGGCTCAG; REV:TGCTGCCTCCCGTAGGAGT	NA	NA	NA	NA	Dan Knights/Antonio	SDSU	GAZ:United States of America	NA	contamination_nicu_FWIRNKE02	0	NA	NA	NA	NA	n	NICU 1	Scott Kelley	Kelley_Newborn_ICU_Study	pyrosequencing	NA	NA	NA	unclassified metagenome	NA	ENVO:anthropogenic habitat	FLX	contamination_nicu_FWIRNKE02	NA	NA	NA	NA	nicu_contamination_data	Contamination in Newborn Intensive Care Units	408169	CCME	NA	NA	NA	386	"DNA was extracted using a lysozyme extraction followed by extraction with a QIAGEN DNEasy Blood and Tissue Extraction Kit. Each amplicon library was constructed by amplifying the16S rRNA gene using the modified 27F and and the 338R primer pair. Primers contained DNA barcode sequences such as those described by Fierer et al 2008, and the recommended 454 adapter sequences. Amplification conditions were exactly as described Fierer et al 2008."	contamination_nicu	NA	n	NA	FWIRNKE02_MBP17	ENVO:surface	16S rRNA	NA	NA	V2	NA	V1-V2	PlasticSide	NA	NA	NA	32.715329	NA	NICU1	Swab_was_taken_from_NICU_baby_open_bedside_plastic_exterior2_NICU1
MBP4.660768	CGTGACAATGTC	CATGCTGCCTCCCGTAGGAGT	NA	NA	NICU 1	BabyBedside	NICU BabyBedside	sink	NA	n	5/28/09	Kelley_nicu_contamination	NA	FWIRNKE02	NA	mimarks-survey	NA	NA	BabyBedside	NA	0	NA	NA	246.888	2/1/09	NA	0	NA	NA	NA	NA	NA	NA	drscottkelley@gmail.com	NA	NA	0.025	1	NA	NA	This study examined contamination of inanimate hospital environments as possible sources of contamination in Newborn Intensive Care Units	NA	NA	-117.157255	y	NA	SDSU	NA	NA	NA	This study examined contamination of inanimate hospital environments as possible sources of contamination in Newborn Intensive Care Units	ENVO:anthropogenic habitat	TCAG	Evidence of widespread opportunistic pathogen contamination in Newborn Intensive Care Units	FWIRNKE02_CGTGACAATGTC	2	Surface contamination swab sample from Newborn Intensive Care Unit assessing bacterial diversity on inanimate hospital environments.	FWD:AGAGTTTGATCCTGGCTCAG; REV:TGCTGCCTCCCGTAGGAGT	NA	NA	NA	NA	Dan Knights/Antonio	SDSU	GAZ:United States of America	NA	contamination_nicu_FWIRNKE02	0	NA	NA	NA	NA	n	NICU 1	Scott Kelley	Kelley_Newborn_ICU_Study	pyrosequencing	NA	NA	NA	unclassified metagenome	NA	ENVO:anthropogenic habitat	FLX	contamination_nicu_FWIRNKE02	NA	NA	NA	NA	nicu_contamination_data	Contamination in Newborn Intensive Care Units	408169	CCME	NA	NA	NA	386	"DNA was extracted using a lysozyme extraction followed by extraction with a QIAGEN DNEasy Blood and Tissue Extraction Kit. Each amplicon library was constructed by amplifying the16S rRNA gene using the modified 27F and and the 338R primer pair. Primers contained DNA barcode sequences such as those described by Fierer et al 2008, and the recommended 454 adapter sequences. Amplification conditions were exactly as described Fierer et al 2008."	contamination_nicu	NA	n	NA	FWIRNKE02_MBP4	ENVO:surface	16S rRNA	NA	NA	V2	NA	V1-V2	PlasticSide	NA	NA	NA	32.715329	NA	NICU1	Swab_was_taken_from_NICU_baby_open_bedside_plastic_exterior_NICU1
MBR13.660762	CTACACAAGCAC	CATGCTGCCTCCCGTAGGAGT	NA	NA	NICU 1	BabyBedside	NICU BabyBedside	sink	NA	n	5/28/09	Kelley_nicu_contamination	NA	FWIRNKE02	NA	mimarks-survey	NA	NA	BabyBedside	NA	0	NA	NA	246.888	2/1/09	NA	0	NA	NA	NA	NA	NA	NA	drscottkelley@gmail.com	NA	NA	0.025	1	NA	NA	This study examined contamination of inanimate hospital environments as possible sources of contamination in Newborn Intensive Care Units	NA	NA	-117.157255	y	NA	SDSU	NA	NA	NA	This study examined contamination of inanimate hospital environments as possible sources of contamination in Newborn Intensive Care Units	ENVO:anthropogenic habitat	TCAG	Evidence of widespread opportunistic pathogen contamination in Newborn Intensive Care Units	FWIRNKE02_CTACACAAGCAC	2	Surface contamination swab sample from Newborn Intensive Care Unit assessing bacterial diversity on inanimate hospital environments.	FWD:AGAGTTTGATCCTGGCTCAG; REV:TGCTGCCTCCCGTAGGAGT	NA	NA	NA	NA	Dan Knights/Antonio	SDSU	GAZ:United States of America	NA	contamination_nicu_FWIRNKE02	0	NA	NA	NA	NA	n	NICU 1	Scott Kelley	Kelley_Newborn_ICU_Study	pyrosequencing	NA	NA	NA	unclassified metagenome	NA	ENVO:anthropogenic habitat	FLX	contamination_nicu_FWIRNKE02	NA	NA	NA	NA	nicu_contamination_data	Contamination in Newborn Intensive Care Units	408169	CCME	NA	NA	NA	386	"DNA was extracted using a lysozyme extraction followed by extraction with a QIAGEN DNEasy Blood and Tissue Extraction Kit. Each amplicon library was constructed by amplifying the16S rRNA gene using the modified 27F and and the 338R primer pair. Primers contained DNA barcode sequences such as those described by Fierer et al 2008, and the recommended 454 adapter sequences. Amplification conditions were exactly as described Fierer et al 2008."	contamination_nicu	NA	n	NA	FWIRNKE02_MBR13	ENVO:surface	16S rRNA	NA	NA	V2	NA	V1-V2	DrawerHandles	NA	NA	NA	32.715329	NA	NICU1	Swab_was_taken_from_NICU_Baby_open_bedside_drawer_handle_beneath_bed_NICU1
MBT14.660752	CTACATCTAAGC	CATGCTGCCTCCCGTAGGAGT	NA	NA	NICU 1	BabyBedside	NICU BabyBedside	sink	NA	n	5/28/09	Kelley_nicu_contamination	NA	FWIRNKE02	NA	mimarks-survey	NA	NA	BabyBedside	NA	0	NA	NA	246.888	2/1/09	NA	0	NA	NA	NA	NA	NA	NA	drscottkelley@gmail.com	NA	NA	0.025	1	NA	NA	This study examined contamination of inanimate hospital environments as possible sources of contamination in Newborn Intensive Care Units	NA	NA	-117.157255	y	NA	SDSU	NA	NA	NA	This study examined contamination of inanimate hospital environments as possible sources of contamination in Newborn Intensive Care Units	ENVO:anthropogenic habitat	TCAG	Evidence of widespread opportunistic pathogen contamination in Newborn Intensive Care Units	FWIRNKE02_CTACATCTAAGC	2	Surface contamination swab sample from Newborn Intensive Care Unit assessing bacterial diversity on inanimate hospital environments.	FWD:AGAGTTTGATCCTGGCTCAG; REV:TGCTGCCTCCCGTAGGAGT	NA	NA	NA	NA	Dan Knights/Antonio	SDSU	GAZ:United States of America	NA	contamination_nicu_FWIRNKE02	0	NA	NA	NA	NA	n	NICU 1	Scott Kelley	Kelley_Newborn_ICU_Study	pyrosequencing	NA	NA	NA	unclassified metagenome	NA	ENVO:anthropogenic habitat	FLX	contamination_nicu_FWIRNKE02	NA	NA	NA	NA	nicu_contamination_data	Contamination in Newborn Intensive Care Units	408169	CCME	NA	NA	NA	386	"DNA was extracted using a lysozyme extraction followed by extraction with a QIAGEN DNEasy Blood and Tissue Extraction Kit. Each amplicon library was constructed by amplifying the16S rRNA gene using the modified 27F and and the 338R primer pair. Primers contained DNA barcode sequences such as those described by Fierer et al 2008, and the recommended 454 adapter sequences. Amplification conditions were exactly as described Fierer et al 2008."	contamination_nicu	NA	n	NA	FWIRNKE02_MBT14	ENVO:surface	16S rRNA	NA	NA	V2	NA	V1-V2	TouchScreen	NA	NA	NA	32.715329	NA	NICU1	Swab_was_taken_from_NICU_baby_open_bedside_touch_screen_on_vitals_monitoring_device_NICU1
UBD22.660753	CTAGTCAGCTGA	CATGCTGCCTCCCGTAGGAGT	NA	NA	NICU 2	BabyBedside	NICU BabyBedside	sink	NA	n	5/28/09	Kelley_nicu_contamination	NA	FWIRNKE02	NA	mimarks-survey	NA	NA	BabyBedside	NA	0	NA	NA	246.888	2/1/09	NA	0	NA	NA	NA	NA	NA	NA	drscottkelley@gmail.com	NA	NA	0.025	1	NA	NA	This study examined contamination of inanimate hospital environments as possible sources of contamination in Newborn Intensive Care Units	NA	NA	-117.157255	y	NA	SDSU	NA	NA	NA	This study examined contamination of inanimate hospital environments as possible sources of contamination in Newborn Intensive Care Units	ENVO:anthropogenic habitat	TCAG	Evidence of widespread opportunistic pathogen contamination in Newborn Intensive Care Units	FWIRNKE02_CTAGTCAGCTGA	2	Surface contamination swab sample from Newborn Intensive Care Unit assessing bacterial diversity on inanimate hospital environments.	FWD:AGAGTTTGATCCTGGCTCAG; REV:TGCTGCCTCCCGTAGGAGT	NA	NA	NA	NA	Dan Knights/Antonio	SDSU	GAZ:United States of America	NA	contamination_nicu_FWIRNKE02	0	NA	NA	NA	NA	n	NICU 2	Scott Kelley	Kelley_Newborn_ICU_Study	pyrosequencing	NA	NA	NA	unclassified metagenome	NA	ENVO:anthropogenic habitat	FLX	contamination_nicu_FWIRNKE02	NA	NA	NA	NA	nicu_contamination_data	Contamination in Newborn Intensive Care Units	408169	CCME	NA	NA	NA	386	"DNA was extracted using a lysozyme extraction followed by extraction with a QIAGEN DNEasy Blood and Tissue Extraction Kit. Each amplicon library was constructed by amplifying the16S rRNA gene using the modified 27F and and the 338R primer pair. Primers contained DNA barcode sequences such as those described by Fierer et al 2008, and the recommended 454 adapter sequences. Amplification conditions were exactly as described Fierer et al 2008."	contamination_nicu	NA	n	NA	FWIRNKE02_UBD22	ENVO:surface	16S rRNA	NA	NA	V2	NA	V1-V2	DiaperCounter	NA	NA	NA	32.715329	NA	NICU 2	Swab_was_taken_from_NICU_baby_open_bedside_diaper_counter_machinE_NICU 2
UBR25.660755	CTATCTAGCGAG	CATGCTGCCTCCCGTAGGAGT	NA	NA	NICU 2	BabyBedside	NICU BabyBedside	sink	NA	n	5/28/09	Kelley_nicu_contamination	NA	FWIRNKE02	NA	mimarks-survey	NA	NA	BabyBedside	NA	0	NA	NA	246.888	2/1/09	NA	0	NA	NA	NA	NA	NA	NA	drscottkelley@gmail.com	NA	NA	0.025	1	NA	NA	This study examined contamination of inanimate hospital environments as possible sources of contamination in Newborn Intensive Care Units	NA	NA	-117.157255	y	NA	SDSU	NA	NA	NA	This study examined contamination of inanimate hospital environments as possible sources of contamination in Newborn Intensive Care Units	ENVO:anthropogenic habitat	TCAG	Evidence of widespread opportunistic pathogen contamination in Newborn Intensive Care Units	FWIRNKE02_CTATCTAGCGAG	2	Surface contamination swab sample from Newborn Intensive Care Unit assessing bacterial diversity on inanimate hospital environments.	FWD:AGAGTTTGATCCTGGCTCAG; REV:TGCTGCCTCCCGTAGGAGT	NA	NA	NA	NA	Dan Knights/Antonio	SDSU	GAZ:United States of America	NA	contamination_nicu_FWIRNKE02	0	NA	NA	NA	NA	n	NICU 2	Scott Kelley	Kelley_Newborn_ICU_Study	pyrosequencing	NA	NA	NA	unclassified metagenome	NA	ENVO:anthropogenic habitat	FLX	contamination_nicu_FWIRNKE02	NA	NA	NA	NA	nicu_contamination_data	Contamination in Newborn Intensive Care Units	408169	CCME	NA	NA	NA	386	"DNA was extracted using a lysozyme extraction followed by extraction with a QIAGEN DNEasy Blood and Tissue Extraction Kit. Each amplicon library was constructed by amplifying the16S rRNA gene using the modified 27F and and the 338R primer pair. Primers contained DNA barcode sequences such as those described by Fierer et al 2008, and the recommended 454 adapter sequences. Amplification conditions were exactly as described Fierer et al 2008."	contamination_nicu	NA	n	NA	FWIRNKE02_UBR25	ENVO:surface	16S rRNA	NA	NA	V2	NA	V1-V2	DrawerHandles	NA	NA	NA	32.715329	NA	NICU 2	Swab_was_taken_from_NICU_Baby_open_bedside_drawer_handle_beneath_bed_NICU 2
UBT30.660772	CTCCACATGAGA	CATGCTGCCTCCCGTAGGAGT	NA	NA	NICU 2	BabyBedside	NICU BabyBedside	sink	NA	n	5/28/09	Kelley_nicu_contamination	NA	FWIRNKE02	NA	mimarks-survey	NA	NA	BabyBedside	NA	0	NA	NA	246.888	2/1/09	NA	0	NA	NA	NA	NA	NA	NA	drscottkelley@gmail.com	NA	NA	0.025	1	NA	NA	This study examined contamination of inanimate hospital environments as possible sources of contamination in Newborn Intensive Care Units	NA	NA	-117.157255	y	NA	SDSU	NA	NA	NA	This study examined contamination of inanimate hospital environments as possible sources of contamination in Newborn Intensive Care Units	ENVO:anthropogenic habitat	TCAG	Evidence of widespread opportunistic pathogen contamination in Newborn Intensive Care Units	FWIRNKE02_CTCCACATGAGA	2	Surface contamination swab sample from Newborn Intensive Care Unit assessing bacterial diversity on inanimate hospital environments.	FWD:AGAGTTTGATCCTGGCTCAG; REV:TGCTGCCTCCCGTAGGAGT	NA	NA	NA	NA	Dan Knights/Antonio	SDSU	GAZ:United States of America	NA	contamination_nicu_FWIRNKE02	0	NA	NA	NA	NA	n	NICU 2	Scott Kelley	Kelley_Newborn_ICU_Study	pyrosequencing	NA	NA	NA	unclassified metagenome	NA	ENVO:anthropogenic habitat	FLX	contamination_nicu_FWIRNKE02	NA	NA	NA	NA	nicu_contamination_data	Contamination in Newborn Intensive Care Units	408169	CCME	NA	NA	NA	386	"DNA was extracted using a lysozyme extraction followed by extraction with a QIAGEN DNEasy Blood and Tissue Extraction Kit. Each amplicon library was constructed by amplifying the16S rRNA gene using the modified 27F and and the 338R primer pair. Primers contained DNA barcode sequences such as those described by Fierer et al 2008, and the recommended 454 adapter sequences. Amplification conditions were exactly as described Fierer et al 2008."	contamination_nicu	NA	n	NA	FWIRNKE02_UBT30	ENVO:surface	16S rRNA	NA	NA	V2	NA	V1-V2	TouchScreen	NA	NA	NA	32.715329	NA	NICU 2	Swab_was_taken_from_NICU_baby_open_bedside_touch_screen_on_vitals_monitoring_device_NICU 2
MDB6.660747	CGTGCATTATCA	CATGCTGCCTCCCGTAGGAGT	NA	NA	NICU 1	DoorButton	NICU DoorButton	sink	NA	n	5/28/09	Kelley_nicu_contamination	NA	FWIRNKE02	NA	mimarks-survey	NA	NA	DoorButton	NA	0	NA	NA	246.888	2/1/09	NA	0	NA	NA	NA	NA	NA	NA	drscottkelley@gmail.com	NA	NA	0.025	1	NA	NA	This study examined contamination of inanimate hospital environments as possible sources of contamination in Newborn Intensive Care Units	NA	NA	-117.157255	y	NA	SDSU	NA	NA	NA	This study examined contamination of inanimate hospital environments as possible sources of contamination in Newborn Intensive Care Units	ENVO:anthropogenic habitat	TCAG	Evidence of widespread opportunistic pathogen contamination in Newborn Intensive Care Units	FWIRNKE02_CGTGCATTATCA	2	Surface contamination swab sample from Newborn Intensive Care Unit assessing bacterial diversity on inanimate hospital environments.	FWD:AGAGTTTGATCCTGGCTCAG; REV:TGCTGCCTCCCGTAGGAGT	NA	NA	NA	NA	Dan Knights/Antonio	SDSU	GAZ:United States of America	NA	contamination_nicu_FWIRNKE02	0	NA	NA	NA	NA	n	NICU 1	Scott Kelley	Kelley_Newborn_ICU_Study	pyrosequencing	NA	NA	NA	unclassified metagenome	NA	ENVO:anthropogenic habitat	FLX	contamination_nicu_FWIRNKE02	NA	NA	NA	NA	nicu_contamination_data	Contamination in Newborn Intensive Care Units	408169	CCME	NA	NA	NA	386	"DNA was extracted using a lysozyme extraction followed by extraction with a QIAGEN DNEasy Blood and Tissue Extraction Kit. Each amplicon library was constructed by amplifying the16S rRNA gene using the modified 27F and and the 338R primer pair. Primers contained DNA barcode sequences such as those described by Fierer et al 2008, and the recommended 454 adapter sequences. Amplification conditions were exactly as described Fierer et al 2008."	contamination_nicu	NA	n	NA	FWIRNKE02_MDB6	ENVO:surface	16S rRNA	NA	NA	V2	NA	V1-V2	ButtonSurface	NA	NA	NA	32.715329	NA	NICU1	Swab_was_taken_from_NICU_door_button_surface1_NICU1
MDB7.660769	CGTGTACATCAG	CATGCTGCCTCCCGTAGGAGT	NA	NA	NICU 1	DoorButton	NICU DoorButton	sink	NA	n	5/28/09	Kelley_nicu_contamination	NA	FWIRNKE02	NA	mimarks-survey	NA	NA	DoorButton	NA	0	NA	NA	246.888	2/1/09	NA	0	NA	NA	NA	NA	NA	NA	drscottkelley@gmail.com	NA	NA	0.025	1	NA	NA	This study examined contamination of inanimate hospital environments as possible sources of contamination in Newborn Intensive Care Units	NA	NA	-117.157255	y	NA	SDSU	NA	NA	NA	This study examined contamination of inanimate hospital environments as possible sources of contamination in Newborn Intensive Care Units	ENVO:anthropogenic habitat	TCAG	Evidence of widespread opportunistic pathogen contamination in Newborn Intensive Care Units	FWIRNKE02_CGTGTACATCAG	2	Surface contamination swab sample from Newborn Intensive Care Unit assessing bacterial diversity on inanimate hospital environments.	FWD:AGAGTTTGATCCTGGCTCAG; REV:TGCTGCCTCCCGTAGGAGT	NA	NA	NA	NA	Dan Knights/Antonio	SDSU	GAZ:United States of America	NA	contamination_nicu_FWIRNKE02	0	NA	NA	NA	NA	n	NICU 1	Scott Kelley	Kelley_Newborn_ICU_Study	pyrosequencing	NA	NA	NA	unclassified metagenome	NA	ENVO:anthropogenic habitat	FLX	contamination_nicu_FWIRNKE02	NA	NA	NA	NA	nicu_contamination_data	Contamination in Newborn Intensive Care Units	408169	CCME	NA	NA	NA	386	"DNA was extracted using a lysozyme extraction followed by extraction with a QIAGEN DNEasy Blood and Tissue Extraction Kit. Each amplicon library was constructed by amplifying the16S rRNA gene using the modified 27F and and the 338R primer pair. Primers contained DNA barcode sequences such as those described by Fierer et al 2008, and the recommended 454 adapter sequences. Amplification conditions were exactly as described Fierer et al 2008."	contamination_nicu	NA	n	NA	FWIRNKE02_MDB7	ENVO:surface	16S rRNA	NA	NA	V2	NA	V1-V2	ButtonSurface	NA	NA	NA	32.715329	NA	NICU1	Swab_was_taken_from_NICU_door_button_surface2_NICU1
UDB19.660770	CTAGAGACTCTT	CATGCTGCCTCCCGTAGGAGT	NA	NA	NICU 2	DoorButton	NICU DoorButton	sink	NA	n	5/28/09	Kelley_nicu_contamination	NA	FWIRNKE02	NA	mimarks-survey	NA	NA	DoorButton	NA	0	NA	NA	246.888	2/1/09	NA	0	NA	NA	NA	NA	NA	NA	drscottkelley@gmail.com	NA	NA	0.025	1	NA	NA	This study examined contamination of inanimate hospital environments as possible sources of contamination in Newborn Intensive Care Units	NA	NA	-117.157255	y	NA	SDSU	NA	NA	NA	This study examined contamination of inanimate hospital environments as possible sources of contamination in Newborn Intensive Care Units	ENVO:anthropogenic habitat	TCAG	Evidence of widespread opportunistic pathogen contamination in Newborn Intensive Care Units	FWIRNKE02_CTAGAGACTCTT	2	Surface contamination swab sample from Newborn Intensive Care Unit assessing bacterial diversity on inanimate hospital environments.	FWD:AGAGTTTGATCCTGGCTCAG; REV:TGCTGCCTCCCGTAGGAGT	NA	NA	NA	NA	Dan Knights/Antonio	SDSU	GAZ:United States of America	NA	contamination_nicu_FWIRNKE02	0	NA	NA	NA	NA	n	NICU 2	Scott Kelley	Kelley_Newborn_ICU_Study	pyrosequencing	NA	NA	NA	unclassified metagenome	NA	ENVO:anthropogenic habitat	FLX	contamination_nicu_FWIRNKE02	NA	NA	NA	NA	nicu_contamination_data	Contamination in Newborn Intensive Care Units	408169	CCME	NA	NA	NA	386	"DNA was extracted using a lysozyme extraction followed by extraction with a QIAGEN DNEasy Blood and Tissue Extraction Kit. Each amplicon library was constructed by amplifying the16S rRNA gene using the modified 27F and and the 338R primer pair. Primers contained DNA barcode sequences such as those described by Fierer et al 2008, and the recommended 454 adapter sequences. Amplification conditions were exactly as described Fierer et al 2008."	contamination_nicu	NA	n	NA	FWIRNKE02_UDB19	ENVO:surface	16S rRNA	NA	NA	V2	NA	V1-V2	ButtonSurface	NA	NA	NA	32.715329	NA	NICU 2	Swab_was_taken_from_NICU_door_button_surface1_NICU 2
UDB31.660751	CTCCTACTGTCT	CATGCTGCCTCCCGTAGGAGT	NA	NA	NICU 2	DoorButton	NICU DoorButton	sink	NA	n	5/28/09	Kelley_nicu_contamination	NA	FWIRNKE02	NA	mimarks-survey	NA	NA	DoorButton	NA	0	NA	NA	246.888	2/1/09	NA	0	NA	NA	NA	NA	NA	NA	drscottkelley@gmail.com	NA	NA	0.025	1	NA	NA	This study examined contamination of inanimate hospital environments as possible sources of contamination in Newborn Intensive Care Units	NA	NA	-117.157255	y	NA	SDSU	NA	NA	NA	This study examined contamination of inanimate hospital environments as possible sources of contamination in Newborn Intensive Care Units	ENVO:anthropogenic habitat	TCAG	Evidence of widespread opportunistic pathogen contamination in Newborn Intensive Care Units	FWIRNKE02_CTCCTACTGTCT	2	Surface contamination swab sample from Newborn Intensive Care Unit assessing bacterial diversity on inanimate hospital environments.	FWD:AGAGTTTGATCCTGGCTCAG; REV:TGCTGCCTCCCGTAGGAGT	NA	NA	NA	NA	Dan Knights/Antonio	SDSU	GAZ:United States of America	NA	contamination_nicu_FWIRNKE02	0	NA	NA	NA	NA	n	NICU 2	Scott Kelley	Kelley_Newborn_ICU_Study	pyrosequencing	NA	NA	NA	unclassified metagenome	NA	ENVO:anthropogenic habitat	FLX	contamination_nicu_FWIRNKE02	NA	NA	NA	NA	nicu_contamination_data	Contamination in Newborn Intensive Care Units	408169	CCME	NA	NA	NA	386	"DNA was extracted using a lysozyme extraction followed by extraction with a QIAGEN DNEasy Blood and Tissue Extraction Kit. Each amplicon library was constructed by amplifying the16S rRNA gene using the modified 27F and and the 338R primer pair. Primers contained DNA barcode sequences such as those described by Fierer et al 2008, and the recommended 454 adapter sequences. Amplification conditions were exactly as described Fierer et al 2008."	contamination_nicu	NA	n	NA	FWIRNKE02_UDB31	ENVO:surface	16S rRNA	NA	NA	V2	NA	V1-V2	ButtonSurface	NA	NA	NA	32.715329	NA	NICU 2	Swab_was_taken_from_NICU_door_button_surface2_NICU 2
MII5.660766	CGTGATCTCTCC	CATGCTGCCTCCCGTAGGAGT	NA	NA	NICU 1	Incubator	NICU Incubator	sink	NA	n	5/28/09	Kelley_nicu_contamination	NA	FWIRNKE02	NA	mimarks-survey	NA	NA	Incubator	NA	0	NA	NA	246.888	2/1/09	NA	0	NA	NA	NA	NA	NA	NA	drscottkelley@gmail.com	NA	NA	0.025	1	NA	NA	This study examined contamination of inanimate hospital environments as possible sources of contamination in Newborn Intensive Care Units	NA	NA	-117.157255	y	NA	SDSU	NA	NA	NA	This study examined contamination of inanimate hospital environments as possible sources of contamination in Newborn Intensive Care Units	ENVO:anthropogenic habitat	TCAG	Evidence of widespread opportunistic pathogen contamination in Newborn Intensive Care Units	FWIRNKE02_CGTGATCTCTCC	2	Surface contamination swab sample from Newborn Intensive Care Unit assessing bacterial diversity on inanimate hospital environments.	FWD:AGAGTTTGATCCTGGCTCAG; REV:TGCTGCCTCCCGTAGGAGT	NA	NA	NA	NA	Dan Knights/Antonio	SDSU	GAZ:United States of America	NA	contamination_nicu_FWIRNKE02	0	NA	NA	NA	NA	n	NICU 1	Scott Kelley	Kelley_Newborn_ICU_Study	pyrosequencing	NA	NA	NA	unclassified metagenome	NA	ENVO:anthropogenic habitat	FLX	contamination_nicu_FWIRNKE02	NA	NA	NA	NA	nicu_contamination_data	Contamination in Newborn Intensive Care Units	408169	CCME	NA	NA	NA	386	"DNA was extracted using a lysozyme extraction followed by extraction with a QIAGEN DNEasy Blood and Tissue Extraction Kit. Each amplicon library was constructed by amplifying the16S rRNA gene using the modified 27F and and the 338R primer pair. Primers contained DNA barcode sequences such as those described by Fierer et al 2008, and the recommended 454 adapter sequences. Amplification conditions were exactly as described Fierer et al 2008."	contamination_nicu	NA	n	NA	FWIRNKE02_MII5	ENVO:surface	16S rRNA	NA	NA	V2	NA	V1-V2	InsideIncubator	NA	NA	NA	32.715329	NA	NICU1	Swab_was_taken_from_NICU_incubator_inside_plastic_NICU1
MIK12.660773	CTAACGCAGTCA	CATGCTGCCTCCCGTAGGAGT	NA	NA	NICU 1	Incubator	NICU Incubator	sink	NA	n	5/28/09	Kelley_nicu_contamination	NA	FWIRNKE02	NA	mimarks-survey	NA	NA	Incubator	NA	0	NA	NA	246.888	2/1/09	NA	0	NA	NA	NA	NA	NA	NA	drscottkelley@gmail.com	NA	NA	0.025	1	NA	NA	This study examined contamination of inanimate hospital environments as possible sources of contamination in Newborn Intensive Care Units	NA	NA	-117.157255	y	NA	SDSU	NA	NA	NA	This study examined contamination of inanimate hospital environments as possible sources of contamination in Newborn Intensive Care Units	ENVO:anthropogenic habitat	TCAG	Evidence of widespread opportunistic pathogen contamination in Newborn Intensive Care Units	FWIRNKE02_CTAACGCAGTCA	2	Surface contamination swab sample from Newborn Intensive Care Unit assessing bacterial diversity on inanimate hospital environments.	FWD:AGAGTTTGATCCTGGCTCAG; REV:TGCTGCCTCCCGTAGGAGT	NA	NA	NA	NA	Dan Knights/Antonio	SDSU	GAZ:United States of America	NA	contamination_nicu_FWIRNKE02	0	NA	NA	NA	NA	n	NICU 1	Scott Kelley	Kelley_Newborn_ICU_Study	pyrosequencing	NA	NA	NA	unclassified metagenome	NA	ENVO:anthropogenic habitat	FLX	contamination_nicu_FWIRNKE02	NA	NA	NA	NA	nicu_contamination_data	Contamination in Newborn Intensive Care Units	408169	CCME	NA	NA	NA	386	"DNA was extracted using a lysozyme extraction followed by extraction with a QIAGEN DNEasy Blood and Tissue Extraction Kit. Each amplicon library was constructed by amplifying the16S rRNA gene using the modified 27F and and the 338R primer pair. Primers contained DNA barcode sequences such as those described by Fierer et al 2008, and the recommended 454 adapter sequences. Amplification conditions were exactly as described Fierer et al 2008."	contamination_nicu	NA	n	NA	FWIRNKE02_MIK12	ENVO:surface	16S rRNA	NA	NA	V2	NA	V1-V2	Keyboard	NA	NA	NA	32.715329	NA	NICU1	Swab_was_taken_from_NICU_incubator_computer_keyboard_NICU1
MIW3.660761	CGTCGATCTCTC	CATGCTGCCTCCCGTAGGAGT	NA	NA	NICU 1	Incubator	NICU Incubator	sink	NA	n	5/28/09	Kelley_nicu_contamination	NA	FWIRNKE02	NA	mimarks-survey	NA	NA	Incubator	NA	0	NA	NA	246.888	2/1/09	NA	0	NA	NA	NA	NA	NA	NA	drscottkelley@gmail.com	NA	NA	0.025	1	NA	NA	This study examined contamination of inanimate hospital environments as possible sources of contamination in Newborn Intensive Care Units	NA	NA	-117.157255	y	NA	SDSU	NA	NA	NA	This study examined contamination of inanimate hospital environments as possible sources of contamination in Newborn Intensive Care Units	ENVO:anthropogenic habitat	TCAG	Evidence of widespread opportunistic pathogen contamination in Newborn Intensive Care Units	FWIRNKE02_CGTCGATCTCTC	2	Surface contamination swab sample from Newborn Intensive Care Unit assessing bacterial diversity on inanimate hospital environments.	FWD:AGAGTTTGATCCTGGCTCAG; REV:TGCTGCCTCCCGTAGGAGT	NA	NA	NA	NA	Dan Knights/Antonio	SDSU	GAZ:United States of America	NA	contamination_nicu_FWIRNKE02	0	NA	NA	NA	NA	n	NICU 1	Scott Kelley	Kelley_Newborn_ICU_Study	pyrosequencing	NA	NA	NA	unclassified metagenome	NA	ENVO:anthropogenic habitat	FLX	contamination_nicu_FWIRNKE02	NA	NA	NA	NA	nicu_contamination_data	Contamination in Newborn Intensive Care Units	408169	CCME	NA	NA	NA	386	"DNA was extracted using a lysozyme extraction followed by extraction with a QIAGEN DNEasy Blood and Tissue Extraction Kit. Each amplicon library was constructed by amplifying the16S rRNA gene using the modified 27F and and the 338R primer pair. Primers contained DNA barcode sequences such as those described by Fierer et al 2008, and the recommended 454 adapter sequences. Amplification conditions were exactly as described Fierer et al 2008."	contamination_nicu	NA	n	NA	FWIRNKE02_MIW3	ENVO:surface	16S rRNA	NA	NA	V2	NA	V1-V2	WhiteTurnHandles	NA	NA	NA	32.715329	NA	NICU1	Swab_was_taken_from_NICU_incubator_white_turn_handles_to_open_incubator_NICU1
UII29.660748	CTCATGTACAGT	CATGCTGCCTCCCGTAGGAGT	NA	NA	NICU 2	Incubator	NICU Incubator	sink	NA	n	5/28/09	Kelley_nicu_contamination	NA	FWIRNKE02	NA	mimarks-survey	NA	NA	Incubator	NA	0	NA	NA	246.888	2/1/09	NA	0	NA	NA	NA	NA	NA	NA	drscottkelley@gmail.com	NA	NA	0.025	1	NA	NA	This study examined contamination of inanimate hospital environments as possible sources of contamination in Newborn Intensive Care Units	NA	NA	-117.157255	y	NA	SDSU	NA	NA	NA	This study examined contamination of inanimate hospital environments as possible sources of contamination in Newborn Intensive Care Units	ENVO:anthropogenic habitat	TCAG	Evidence of widespread opportunistic pathogen contamination in Newborn Intensive Care Units	FWIRNKE02_CTCATGTACAGT	2	Surface contamination swab sample from Newborn Intensive Care Unit assessing bacterial diversity on inanimate hospital environments.	FWD:AGAGTTTGATCCTGGCTCAG; REV:TGCTGCCTCCCGTAGGAGT	NA	NA	NA	NA	Dan Knights/Antonio	SDSU	GAZ:United States of America	NA	contamination_nicu_FWIRNKE02	0	NA	NA	NA	NA	n	NICU 2	Scott Kelley	Kelley_Newborn_ICU_Study	pyrosequencing	NA	NA	NA	unclassified metagenome	NA	ENVO:anthropogenic habitat	FLX	contamination_nicu_FWIRNKE02	NA	NA	NA	NA	nicu_contamination_data	Contamination in Newborn Intensive Care Units	408169	CCME	NA	NA	NA	386	"DNA was extracted using a lysozyme extraction followed by extraction with a QIAGEN DNEasy Blood and Tissue Extraction Kit. Each amplicon library was constructed by amplifying the16S rRNA gene using the modified 27F and and the 338R primer pair. Primers contained DNA barcode sequences such as those described by Fierer et al 2008, and the recommended 454 adapter sequences. Amplification conditions were exactly as described Fierer et al 2008."	contamination_nicu	NA	n	NA	FWIRNKE02_UII29	ENVO:surface	16S rRNA	NA	NA	V2	NA	V1-V2	InsideIncubator	NA	NA	NA	32.715329	NA	NICU 2	Swab_was_taken_from_NICU_incubator_inside_plastic_NICU 2
UIK27.660759	CTCAATGACTCA	CATGCTGCCTCCCGTAGGAGT	NA	NA	NICU 2	Incubator	NICU Incubator	sink	NA	n	5/28/09	Kelley_nicu_contamination	NA	FWIRNKE02	NA	mimarks-survey	NA	NA	Incubator	NA	0	NA	NA	246.888	2/1/09	NA	0	NA	NA	NA	NA	NA	NA	drscottkelley@gmail.com	NA	NA	0.025	1	NA	NA	This study examined contamination of inanimate hospital environments as possible sources of contamination in Newborn Intensive Care Units	NA	NA	-117.157255	y	NA	SDSU	NA	NA	NA	This study examined contamination of inanimate hospital environments as possible sources of contamination in Newborn Intensive Care Units	ENVO:anthropogenic habitat	TCAG	Evidence of widespread opportunistic pathogen contamination in Newborn Intensive Care Units	FWIRNKE02_CTCAATGACTCA	2	Surface contamination swab sample from Newborn Intensive Care Unit assessing bacterial diversity on inanimate hospital environments.	FWD:AGAGTTTGATCCTGGCTCAG; REV:TGCTGCCTCCCGTAGGAGT	NA	NA	NA	NA	Dan Knights/Antonio	SDSU	GAZ:United States of America	NA	contamination_nicu_FWIRNKE02	0	NA	NA	NA	NA	n	NICU 2	Scott Kelley	Kelley_Newborn_ICU_Study	pyrosequencing	NA	NA	NA	unclassified metagenome	NA	ENVO:anthropogenic habitat	FLX	contamination_nicu_FWIRNKE02	NA	NA	NA	NA	nicu_contamination_data	Contamination in Newborn Intensive Care Units	408169	CCME	NA	NA	NA	386	"DNA was extracted using a lysozyme extraction followed by extraction with a QIAGEN DNEasy Blood and Tissue Extraction Kit. Each amplicon library was constructed by amplifying the16S rRNA gene using the modified 27F and and the 338R primer pair. Primers contained DNA barcode sequences such as those described by Fierer et al 2008, and the recommended 454 adapter sequences. Amplification conditions were exactly as described Fierer et al 2008."	contamination_nicu	NA	n	NA	FWIRNKE02_UIK27	ENVO:surface	16S rRNA	NA	NA	V2	NA	V1-V2	Keyboard	NA	NA	NA	32.715329	NA	NICU 2	Swab_was_taken_from_NICU_incubator_computer_keyboard_NICU 2
UIR24.660760	CTATCAGTGTAC	CATGCTGCCTCCCGTAGGAGT	NA	NA	NICU 2	Incubator	NICU Incubator	sink	NA	n	5/28/09	Kelley_nicu_contamination	NA	FWIRNKE02	NA	mimarks-survey	NA	NA	Incubator	NA	0	NA	NA	246.888	2/1/09	NA	0	NA	NA	NA	NA	NA	NA	drscottkelley@gmail.com	NA	NA	0.025	1	NA	NA	This study examined contamination of inanimate hospital environments as possible sources of contamination in Newborn Intensive Care Units	NA	NA	-117.157255	y	NA	SDSU	NA	NA	NA	This study examined contamination of inanimate hospital environments as possible sources of contamination in Newborn Intensive Care Units	ENVO:anthropogenic habitat	TCAG	Evidence of widespread opportunistic pathogen contamination in Newborn Intensive Care Units	FWIRNKE02_CTATCAGTGTAC	2	Surface contamination swab sample from Newborn Intensive Care Unit assessing bacterial diversity on inanimate hospital environments.	FWD:AGAGTTTGATCCTGGCTCAG; REV:TGCTGCCTCCCGTAGGAGT	NA	NA	NA	NA	Dan Knights/Antonio	SDSU	GAZ:United States of America	NA	contamination_nicu_FWIRNKE02	0	NA	NA	NA	NA	n	NICU 2	Scott Kelley	Kelley_Newborn_ICU_Study	pyrosequencing	NA	NA	NA	unclassified metagenome	NA	ENVO:anthropogenic habitat	FLX	contamination_nicu_FWIRNKE02	NA	NA	NA	NA	nicu_contamination_data	Contamination in Newborn Intensive Care Units	408169	CCME	NA	NA	NA	386	"DNA was extracted using a lysozyme extraction followed by extraction with a QIAGEN DNEasy Blood and Tissue Extraction Kit. Each amplicon library was constructed by amplifying the16S rRNA gene using the modified 27F and and the 338R primer pair. Primers contained DNA barcode sequences such as those described by Fierer et al 2008, and the recommended 454 adapter sequences. Amplification conditions were exactly as described Fierer et al 2008."	contamination_nicu	NA	n	NA	FWIRNKE02_UIR24	ENVO:surface	16S rRNA	NA	NA	V2	NA	V1-V2	DrawerHandles	NA	NA	NA	32.715329	NA	NICU 2	Swab_was_taken_from_NICU_incubator_drawer_handle_beneath_incubator_NICU 2
UIW23.660774	CTATAGTCGTGT	CATGCTGCCTCCCGTAGGAGT	NA	NA	NICU 2	Incubator	NICU Incubator	sink	NA	n	5/28/09	Kelley_nicu_contamination	NA	FWIRNKE02	NA	mimarks-survey	NA	NA	Incubator	NA	0	NA	NA	246.888	2/1/09	NA	0	NA	NA	NA	NA	NA	NA	drscottkelley@gmail.com	NA	NA	0.025	1	NA	NA	This study examined contamination of inanimate hospital environments as possible sources of contamination in Newborn Intensive Care Units	NA	NA	-117.157255	y	NA	SDSU	NA	NA	NA	This study examined contamination of inanimate hospital environments as possible sources of contamination in Newborn Intensive Care Units	ENVO:anthropogenic habitat	TCAG	Evidence of widespread opportunistic pathogen contamination in Newborn Intensive Care Units	FWIRNKE02_CTATAGTCGTGT	2	Surface contamination swab sample from Newborn Intensive Care Unit assessing bacterial diversity on inanimate hospital environments.	FWD:AGAGTTTGATCCTGGCTCAG; REV:TGCTGCCTCCCGTAGGAGT	NA	NA	NA	NA	Dan Knights/Antonio	SDSU	GAZ:United States of America	NA	contamination_nicu_FWIRNKE02	0	NA	NA	NA	NA	n	NICU 2	Scott Kelley	Kelley_Newborn_ICU_Study	pyrosequencing	NA	NA	NA	unclassified metagenome	NA	ENVO:anthropogenic habitat	FLX	contamination_nicu_FWIRNKE02	NA	NA	NA	NA	nicu_contamination_data	Contamination in Newborn Intensive Care Units	408169	CCME	NA	NA	NA	386	"DNA was extracted using a lysozyme extraction followed by extraction with a QIAGEN DNEasy Blood and Tissue Extraction Kit. Each amplicon library was constructed by amplifying the16S rRNA gene using the modified 27F and and the 338R primer pair. Primers contained DNA barcode sequences such as those described by Fierer et al 2008, and the recommended 454 adapter sequences. Amplification conditions were exactly as described Fierer et al 2008."	contamination_nicu	NA	n	NA	FWIRNKE02_UIW23	ENVO:surface	16S rRNA	NA	NA	V2	NA	V1-V2	WhiteTurnHandles	NA	NA	NA	32.715329	NA	NICU 2	Swab_was_taken_from_NICU_incubator_white_turn_handles_to_open_incubator_NICU 2
MPK2.660771	CGTCAGACGGAT	CATGCTGCCTCCCGTAGGAGT	NA	NA	NICU 1	Pyxus	NICU Pyxus	sink	NA	n	5/28/09	Kelley_nicu_contamination	NA	FWIRNKE02	NA	mimarks-survey	NA	NA	Pyxus	NA	0	NA	NA	246.888	2/1/09	NA	0	NA	NA	NA	NA	NA	NA	drscottkelley@gmail.com	NA	NA	0.025	1	NA	NA	This study examined contamination of inanimate hospital environments as possible sources of contamination in Newborn Intensive Care Units	NA	NA	-117.157255	y	NA	SDSU	NA	NA	NA	This study examined contamination of inanimate hospital environments as possible sources of contamination in Newborn Intensive Care Units	ENVO:anthropogenic habitat	TCAG	Evidence of widespread opportunistic pathogen contamination in Newborn Intensive Care Units	FWIRNKE02_CGTCAGACGGAT	2	Surface contamination swab sample from Newborn Intensive Care Unit assessing bacterial diversity on inanimate hospital environments.	FWD:AGAGTTTGATCCTGGCTCAG; REV:TGCTGCCTCCCGTAGGAGT	NA	NA	NA	NA	Dan Knights/Antonio	SDSU	GAZ:United States of America	NA	contamination_nicu_FWIRNKE02	0	NA	NA	NA	NA	n	NICU 1	Scott Kelley	Kelley_Newborn_ICU_Study	pyrosequencing	NA	NA	NA	unclassified metagenome	NA	ENVO:anthropogenic habitat	FLX	contamination_nicu_FWIRNKE02	NA	NA	NA	NA	nicu_contamination_data	Contamination in Newborn Intensive Care Units	408169	CCME	NA	NA	NA	386	"DNA was extracted using a lysozyme extraction followed by extraction with a QIAGEN DNEasy Blood and Tissue Extraction Kit. Each amplicon library was constructed by amplifying the16S rRNA gene using the modified 27F and and the 338R primer pair. Primers contained DNA barcode sequences such as those described by Fierer et al 2008, and the recommended 454 adapter sequences. Amplification conditions were exactly as described Fierer et al 2008."	contamination_nicu	NA	n	NA	FWIRNKE02_MPK2	ENVO:surface	16S rRNA	NA	NA	V2	NA	V1-V2	Keyboard	NA	NA	NA	32.715329	NA	NICU1	Swab_was_taken_from_NICU_Pyxus_Med_machine_keyboard_NICU1
UPK18.660765	CTAGAACGCACT	CATGCTGCCTCCCGTAGGAGT	NA	NA	NICU 2	Pyxus	NICU Pyxus	sink	NA	n	5/28/09	Kelley_nicu_contamination	NA	FWIRNKE02	NA	mimarks-survey	NA	NA	Pyxus	NA	0	NA	NA	246.888	2/1/09	NA	0	NA	NA	NA	NA	NA	NA	drscottkelley@gmail.com	NA	NA	0.025	1	NA	NA	This study examined contamination of inanimate hospital environments as possible sources of contamination in Newborn Intensive Care Units	NA	NA	-117.157255	y	NA	SDSU	NA	NA	NA	This study examined contamination of inanimate hospital environments as possible sources of contamination in Newborn Intensive Care Units	ENVO:anthropogenic habitat	TCAG	Evidence of widespread opportunistic pathogen contamination in Newborn Intensive Care Units	FWIRNKE02_CTAGAACGCACT	2	Surface contamination swab sample from Newborn Intensive Care Unit assessing bacterial diversity on inanimate hospital environments.	FWD:AGAGTTTGATCCTGGCTCAG; REV:TGCTGCCTCCCGTAGGAGT	NA	NA	NA	NA	Dan Knights/Antonio	SDSU	GAZ:United States of America	NA	contamination_nicu_FWIRNKE02	0	NA	NA	NA	NA	n	NICU 2	Scott Kelley	Kelley_Newborn_ICU_Study	pyrosequencing	NA	NA	NA	unclassified metagenome	NA	ENVO:anthropogenic habitat	FLX	contamination_nicu_FWIRNKE02	NA	NA	NA	NA	nicu_contamination_data	Contamination in Newborn Intensive Care Units	408169	CCME	NA	NA	NA	386	"DNA was extracted using a lysozyme extraction followed by extraction with a QIAGEN DNEasy Blood and Tissue Extraction Kit. Each amplicon library was constructed by amplifying the16S rRNA gene using the modified 27F and and the 338R primer pair. Primers contained DNA barcode sequences such as those described by Fierer et al 2008, and the recommended 454 adapter sequences. Amplification conditions were exactly as described Fierer et al 2008."	contamination_nicu	NA	n	NA	FWIRNKE02_UPK18	ENVO:surface	16S rRNA	NA	NA	V2	NA	V1-V2	Keyboard	NA	NA	NA	32.715329	NA	NICU 2	Swab_was_taken_from_NICU_Pyxus_Med_machine_keyboard_NICU 2
MSC8.660757	CGTGTGATCAGG	CATGCTGCCTCCCGTAGGAGT	NA	NA	NICU 1	Sink	NICU Sink	sink	NA	n	5/28/09	Kelley_nicu_contamination	NA	FWIRNKE02	NA	mimarks-survey	NA	NA	Sink	NA	0	NA	NA	246.888	2/1/09	NA	0	NA	NA	NA	NA	NA	NA	drscottkelley@gmail.com	NA	NA	0.025	1	NA	NA	This study examined contamination of inanimate hospital environments as possible sources of contamination in Newborn Intensive Care Units	NA	NA	-117.157255	y	NA	SDSU	NA	NA	NA	This study examined contamination of inanimate hospital environments as possible sources of contamination in Newborn Intensive Care Units	ENVO:anthropogenic habitat	TCAG	Evidence of widespread opportunistic pathogen contamination in Newborn Intensive Care Units	FWIRNKE02_CGTGTGATCAGG	2	Surface contamination swab sample from Newborn Intensive Care Unit assessing bacterial diversity on inanimate hospital environments.	FWD:AGAGTTTGATCCTGGCTCAG; REV:TGCTGCCTCCCGTAGGAGT	NA	NA	NA	NA	Dan Knights/Antonio	SDSU	GAZ:United States of America	NA	contamination_nicu_FWIRNKE02	0	NA	NA	NA	NA	n	NICU 1	Scott Kelley	Kelley_Newborn_ICU_Study	pyrosequencing	NA	NA	NA	unclassified metagenome	NA	ENVO:anthropogenic habitat	FLX	contamination_nicu_FWIRNKE02	NA	NA	NA	NA	nicu_contamination_data	Contamination in Newborn Intensive Care Units	408169	CCME	NA	NA	NA	386	"DNA was extracted using a lysozyme extraction followed by extraction with a QIAGEN DNEasy Blood and Tissue Extraction Kit. Each amplicon library was constructed by amplifying the16S rRNA gene using the modified 27F and and the 338R primer pair. Primers contained DNA barcode sequences such as those described by Fierer et al 2008, and the recommended 454 adapter sequences. Amplification conditions were exactly as described Fierer et al 2008."	contamination_nicu	NA	n	NA	FWIRNKE02_MSC8	ENVO:surface	16S rRNA	NA	NA	V2	NA	V1-V2	CabinetHandles	NA	NA	NA	32.715329	NA	NICU1	Swab_was_taken_from_NICU_sink_cabinet_handles_NICU1
MSS1.660763	CGTCACGACTAA	CATGCTGCCTCCCGTAGGAGT	NA	NA	NICU 1	Sink	NICU Sink	sink	NA	n	5/28/09	Kelley_nicu_contamination	NA	FWIRNKE02	NA	mimarks-survey	NA	NA	Sink	NA	0	NA	NA	246.888	2/1/09	NA	0	NA	NA	NA	NA	NA	NA	drscottkelley@gmail.com	NA	NA	0.025	1	NA	NA	This study examined contamination of inanimate hospital environments as possible sources of contamination in Newborn Intensive Care Units	NA	NA	-117.157255	y	NA	SDSU	NA	NA	NA	This study examined contamination of inanimate hospital environments as possible sources of contamination in Newborn Intensive Care Units	ENVO:anthropogenic habitat	TCAG	Evidence of widespread opportunistic pathogen contamination in Newborn Intensive Care Units	FWIRNKE02_CGTCACGACTAA	2	Surface contamination swab sample from Newborn Intensive Care Unit assessing bacterial diversity on inanimate hospital environments.	FWD:AGAGTTTGATCCTGGCTCAG; REV:TGCTGCCTCCCGTAGGAGT	NA	NA	NA	NA	Dan Knights/Antonio	SDSU	GAZ:United States of America	NA	contamination_nicu_FWIRNKE02	0	NA	NA	NA	NA	n	NICU 1	Scott Kelley	Kelley_Newborn_ICU_Study	pyrosequencing	NA	NA	NA	unclassified metagenome	NA	ENVO:anthropogenic habitat	FLX	contamination_nicu_FWIRNKE02	NA	NA	NA	NA	nicu_contamination_data	Contamination in Newborn Intensive Care Units	408169	CCME	NA	NA	NA	386	"DNA was extracted using a lysozyme extraction followed by extraction with a QIAGEN DNEasy Blood and Tissue Extraction Kit. Each amplicon library was constructed by amplifying the16S rRNA gene using the modified 27F and and the 338R primer pair. Primers contained DNA barcode sequences such as those described by Fierer et al 2008, and the recommended 454 adapter sequences. Amplification conditions were exactly as described Fierer et al 2008."	contamination_nicu	NA	n	NA	FWIRNKE02_MSS1	ENVO:surface	16S rRNA	NA	NA	V2	NA	V1-V2	SinkHandle	NA	NA	NA	32.715329	NA	NICU1	Swab_was_taken_from_NICU_sink_handles_NICU1
USO26.660754	CTATGCTTGATG	CATGCTGCCTCCCGTAGGAGT	NA	NA	NICU 2	Sink	NICU Sink	sink	NA	n	5/28/09	Kelley_nicu_contamination	NA	FWIRNKE02	NA	mimarks-survey	NA	NA	Sink	NA	0	NA	NA	246.888	2/1/09	NA	0	NA	NA	NA	NA	NA	NA	drscottkelley@gmail.com	NA	NA	0.025	1	NA	NA	This study examined contamination of inanimate hospital environments as possible sources of contamination in Newborn Intensive Care Units	NA	NA	-117.157255	y	NA	SDSU	NA	NA	NA	This study examined contamination of inanimate hospital environments as possible sources of contamination in Newborn Intensive Care Units	ENVO:anthropogenic habitat	TCAG	Evidence of widespread opportunistic pathogen contamination in Newborn Intensive Care Units	FWIRNKE02_CTATGCTTGATG	2	Surface contamination swab sample from Newborn Intensive Care Unit assessing bacterial diversity on inanimate hospital environments.	FWD:AGAGTTTGATCCTGGCTCAG; REV:TGCTGCCTCCCGTAGGAGT	NA	NA	NA	NA	Dan Knights/Antonio	SDSU	GAZ:United States of America	NA	contamination_nicu_FWIRNKE02	0	NA	NA	NA	NA	n	NICU 2	Scott Kelley	Kelley_Newborn_ICU_Study	pyrosequencing	NA	NA	NA	unclassified metagenome	NA	ENVO:anthropogenic habitat	FLX	contamination_nicu_FWIRNKE02	NA	NA	NA	NA	nicu_contamination_data	Contamination in Newborn Intensive Care Units	408169	CCME	NA	NA	NA	386	"DNA was extracted using a lysozyme extraction followed by extraction with a QIAGEN DNEasy Blood and Tissue Extraction Kit. Each amplicon library was constructed by amplifying the16S rRNA gene using the modified 27F and and the 338R primer pair. Primers contained DNA barcode sequences such as those described by Fierer et al 2008, and the recommended 454 adapter sequences. Amplification conditions were exactly as described Fierer et al 2008."	contamination_nicu	NA	n	NA	FWIRNKE02_USO26	ENVO:surface	16S rRNA	NA	NA	V2	NA	V1-V2	CounterbySink	NA	NA	NA	32.715329	NA	NICU 2	Swab_was_taken_from_NICU_counter_next_to_sink_NICU 2
MWR15.660764	CTACGCGTCTCT	CATGCTGCCTCCCGTAGGAGT	NA	NA	NICU 1	WeighCart	NICU WeighCart	sink	NA	n	5/28/09	Kelley_nicu_contamination	NA	FWIRNKE02	NA	mimarks-survey	NA	NA	WeighCart	NA	0	NA	NA	246.888	2/1/09	NA	0	NA	NA	NA	NA	NA	NA	drscottkelley@gmail.com	NA	NA	0.025	1	NA	NA	This study examined contamination of inanimate hospital environments as possible sources of contamination in Newborn Intensive Care Units	NA	NA	-117.157255	y	NA	SDSU	NA	NA	NA	This study examined contamination of inanimate hospital environments as possible sources of contamination in Newborn Intensive Care Units	ENVO:anthropogenic habitat	TCAG	Evidence of widespread opportunistic pathogen contamination in Newborn Intensive Care Units	FWIRNKE02_CTACGCGTCTCT	2	Surface contamination swab sample from Newborn Intensive Care Unit assessing bacterial diversity on inanimate hospital environments.	FWD:AGAGTTTGATCCTGGCTCAG; REV:TGCTGCCTCCCGTAGGAGT	NA	NA	NA	NA	Dan Knights/Antonio	SDSU	GAZ:United States of America	NA	contamination_nicu_FWIRNKE02	0	NA	NA	NA	NA	n	NICU 1	Scott Kelley	Kelley_Newborn_ICU_Study	pyrosequencing	NA	NA	NA	unclassified metagenome	NA	ENVO:anthropogenic habitat	FLX	contamination_nicu_FWIRNKE02	NA	NA	NA	NA	nicu_contamination_data	Contamination in Newborn Intensive Care Units	408169	CCME	NA	NA	NA	386	"DNA was extracted using a lysozyme extraction followed by extraction with a QIAGEN DNEasy Blood and Tissue Extraction Kit. Each amplicon library was constructed by amplifying the16S rRNA gene using the modified 27F and and the 338R primer pair. Primers contained DNA barcode sequences such as those described by Fierer et al 2008, and the recommended 454 adapter sequences. Amplification conditions were exactly as described Fierer et al 2008."	contamination_nicu	NA	n	NA	FWIRNKE02_MWR15	ENVO:surface	16S rRNA	NA	NA	V2	NA	V1-V2	DrawerHandles	NA	NA	NA	32.715329	NA	NICU1	Swab_was_taken_from_NICU_traveling_weigh_cart_drawer_handles1_NICU1
MWR16.660756	CTACTACAGGTG	CATGCTGCCTCCCGTAGGAGT	NA	NA	NICU 1	WeighCart	NICU WeighCart	sink	NA	n	5/28/09	Kelley_nicu_contamination	NA	FWIRNKE02	NA	mimarks-survey	NA	NA	WeighCart	NA	0	NA	NA	246.888	2/1/09	NA	0	NA	NA	NA	NA	NA	NA	drscottkelley@gmail.com	NA	NA	0.025	1	NA	NA	This study examined contamination of inanimate hospital environments as possible sources of contamination in Newborn Intensive Care Units	NA	NA	-117.157255	y	NA	SDSU	NA	NA	NA	This study examined contamination of inanimate hospital environments as possible sources of contamination in Newborn Intensive Care Units	ENVO:anthropogenic habitat	TCAG	Evidence of widespread opportunistic pathogen contamination in Newborn Intensive Care Units	FWIRNKE02_CTACTACAGGTG	2	Surface contamination swab sample from Newborn Intensive Care Unit assessing bacterial diversity on inanimate hospital environments.	FWD:AGAGTTTGATCCTGGCTCAG; REV:TGCTGCCTCCCGTAGGAGT	NA	NA	NA	NA	Dan Knights/Antonio	SDSU	GAZ:United States of America	NA	contamination_nicu_FWIRNKE02	0	NA	NA	NA	NA	n	NICU 1	Scott Kelley	Kelley_Newborn_ICU_Study	pyrosequencing	NA	NA	NA	unclassified metagenome	NA	ENVO:anthropogenic habitat	FLX	contamination_nicu_FWIRNKE02	NA	NA	NA	NA	nicu_contamination_data	Contamination in Newborn Intensive Care Units	408169	CCME	NA	NA	NA	386	"DNA was extracted using a lysozyme extraction followed by extraction with a QIAGEN DNEasy Blood and Tissue Extraction Kit. Each amplicon library was constructed by amplifying the16S rRNA gene using the modified 27F and and the 338R primer pair. Primers contained DNA barcode sequences such as those described by Fierer et al 2008, and the recommended 454 adapter sequences. Amplification conditions were exactly as described Fierer et al 2008."	contamination_nicu	NA	n	NA	FWIRNKE02_MWR16	ENVO:surface	16S rRNA	NA	NA	V2	NA	V1-V2	DrawerHandles	NA	NA	NA	32.715329	NA	NICU1	Swab_was_taken_from_NICU_traveling_weigh_cart_drawer_handles2_NICU1
UWR20.660776	CTAGCGAACATC	CATGCTGCCTCCCGTAGGAGT	NA	NA	NICU 2	WeighCart	NICU WeighCart	sink	NA	n	5/28/09	Kelley_nicu_contamination	NA	FWIRNKE02	NA	mimarks-survey	NA	NA	WeighCart	NA	0	NA	NA	246.888	2/1/09	NA	0	NA	NA	NA	NA	NA	NA	drscottkelley@gmail.com	NA	NA	0.025	1	NA	NA	This study examined contamination of inanimate hospital environments as possible sources of contamination in Newborn Intensive Care Units	NA	NA	-117.157255	y	NA	SDSU	NA	NA	NA	This study examined contamination of inanimate hospital environments as possible sources of contamination in Newborn Intensive Care Units	ENVO:anthropogenic habitat	TCAG	Evidence of widespread opportunistic pathogen contamination in Newborn Intensive Care Units	FWIRNKE02_CTAGCGAACATC	2	Surface contamination swab sample from Newborn Intensive Care Unit assessing bacterial diversity on inanimate hospital environments.	FWD:AGAGTTTGATCCTGGCTCAG; REV:TGCTGCCTCCCGTAGGAGT	NA	NA	NA	NA	Dan Knights/Antonio	SDSU	GAZ:United States of America	NA	contamination_nicu_FWIRNKE02	0	NA	NA	NA	NA	n	NICU 2	Scott Kelley	Kelley_Newborn_ICU_Study	pyrosequencing	NA	NA	NA	unclassified metagenome	NA	ENVO:anthropogenic habitat	FLX	contamination_nicu_FWIRNKE02	NA	NA	NA	NA	nicu_contamination_data	Contamination in Newborn Intensive Care Units	408169	CCME	NA	NA	NA	386	"DNA was extracted using a lysozyme extraction followed by extraction with a QIAGEN DNEasy Blood and Tissue Extraction Kit. Each amplicon library was constructed by amplifying the16S rRNA gene using the modified 27F and and the 338R primer pair. Primers contained DNA barcode sequences such as those described by Fierer et al 2008, and the recommended 454 adapter sequences. Amplification conditions were exactly as described Fierer et al 2008."	contamination_nicu	NA	n	NA	FWIRNKE02_UWR20	ENVO:surface	16S rRNA	NA	NA	V2	NA	V1-V2	DrawerHandles	NA	NA	NA	32.715329	NA	NICU 2	Swab_was_taken_from_NICU_traveling_weigh_cart_drawer_handles1_NICU 2
UWR21.660749	CTAGGTCACTAG	CATGCTGCCTCCCGTAGGAGT	NA	NA	NICU 2	WeighCart	NICU WeighCart	sink	NA	n	5/28/09	Kelley_nicu_contamination	NA	FWIRNKE02	NA	mimarks-survey	NA	NA	WeighCart	NA	0	NA	NA	246.888	2/1/09	NA	0	NA	NA	NA	NA	NA	NA	drscottkelley@gmail.com	NA	NA	0.025	1	NA	NA	This study examined contamination of inanimate hospital environments as possible sources of contamination in Newborn Intensive Care Units	NA	NA	-117.157255	y	NA	SDSU	NA	NA	NA	This study examined contamination of inanimate hospital environments as possible sources of contamination in Newborn Intensive Care Units	ENVO:anthropogenic habitat	TCAG	Evidence of widespread opportunistic pathogen contamination in Newborn Intensive Care Units	FWIRNKE02_CTAGGTCACTAG	2	Surface contamination swab sample from Newborn Intensive Care Unit assessing bacterial diversity on inanimate hospital environments.	FWD:AGAGTTTGATCCTGGCTCAG; REV:TGCTGCCTCCCGTAGGAGT	NA	NA	NA	NA	Dan Knights/Antonio	SDSU	GAZ:United States of America	NA	contamination_nicu_FWIRNKE02	0	NA	NA	NA	NA	n	NICU 2	Scott Kelley	Kelley_Newborn_ICU_Study	pyrosequencing	NA	NA	NA	unclassified metagenome	NA	ENVO:anthropogenic habitat	FLX	contamination_nicu_FWIRNKE02	NA	NA	NA	NA	nicu_contamination_data	Contamination in Newborn Intensive Care Units	408169	CCME	NA	NA	NA	386	"DNA was extracted using a lysozyme extraction followed by extraction with a QIAGEN DNEasy Blood and Tissue Extraction Kit. Each amplicon library was constructed by amplifying the16S rRNA gene using the modified 27F and and the 338R primer pair. Primers contained DNA barcode sequences such as those described by Fierer et al 2008, and the recommended 454 adapter sequences. Amplification conditions were exactly as described Fierer et al 2008."	contamination_nicu	NA	n	NA	FWIRNKE02_UWR21	ENVO:surface	16S rRNA	NA	NA	V2	NA	V1-V2	DrawerHandles	NA	NA	NA	32.715329	NA	NICU 2	Swab_was_taken_from_NICU_traveling_weigh_cart_drawer_handles2_NICU 2
NYFC16.522479	TCCTAGCAGTGA	CATGCTGCCTCCCGTAGGAGT	NA	NA	NewYork	Chair	NA	NA	Chair	y	2011	Kelley_Office_Map	NA	FWIRNKE01	NA	mimarks-survey	NA	NA	NA	NA	0	NA	NA	9.777	2011	NA	0	NA	NA	female	NA	NA	NA	drscottkelley@gmail.com	NA	16	NA	".1,g"	NA	NA	"Samples from office surfaces in New York, San Francisco and Tucson"	NA	NA	-74.00597	y	NA	CCME	NA	NA	NA	"People in developed countries spend approximately 90% of their lives indoors, yet we know little about the source and diversity of microbes in built environments. In this study, we combined culture-based cell counting and multiplexed pyrosequencing of environmental ribosomal RNA (rRNA) gene sequences to investigate office space bacterial diversity in three metropolitan areas. Five surfaces common to all offices were sampled using sterile double-tipped swabs, one tip for culturing and one for DNA extraction, in 30 different offices per city (90 offices, 450 total samples). 16S rRNA gene sequences were PCR amplified using bar-coded universal bacterial primers from 54 of the surfaces (18 per city) and pooled for pyrosequencing. A three-factorial Analysis of Variance (ANOVA) found significant differences in viable bacterial abundance between offices inhabited by men or women, among the various surface types, and among cities. Multiplex pyrosequencing identified more than 500 bacterial genera from 20 different bacterial divisions. The most abundant of these genera tended to be common inhabitants of human skin, nasal, oral or intestinal cavities. Other commonly occurring genera appeared to have environmental origins (e.g., soils). There were no significant differences in the bacterial diversity between offices inhabited by men or women or among surfaces, but the bacterial community diversity of the Tucson samples was clearly distinguishable from that of New York and San Francisco, which were indistinguishable. Overall, our comprehensive molecular analysis of office building microbial diversity shows the potential of these methods for studying patterns and origins of indoor bacterial contamination. Humans move through a sea of microbial life that is seldom perceived except in the context of potential disease and decay Feazel et al. (2009)."	ENVO:surface	NA	Office Space Bacterial Abundance and Diversity in Three  Metropolitan Areas	NA	NA	Office Map study	FWD:AGAGTTTGATCCTGGCTCAG; REV:TGCTGCCTCCCGTAGGAGT	22666400	NA	NA	NA	Dan Knights/Justin Kuczinsky	CCME	GAZ:United States of America	NA	NA	0	NA	NA	NA	NA	n	NA	Scott Kelley	Kelley_office_contamination	pyrosequencing	NA	NA	NA	metagenomes	NA	ENVO:terrestrial biome	FLX	NA	NA	NA	NA	NA	Kelley_office_contamination	Kelley_Office_Map	408169	CCME	NA	NA	NA	1479	FLX protocol	NA	NA	n	NA	NA	ENVO:surface	16S rRNA	NA	NA	V2	CCME	NA	Chair	NA	NA	NA	40.71435	NA	NA	Swab_sample_from_woman_chair_in_New_York_office_room_16
NYFC22.522465	TCCGTCGTCTGT	CATGCTGCCTCCCGTAGGAGT	NA	NA	NewYork	Chair	NA	NA	Chair	y	2011	Kelley_Office_Map	NA	FWIRNKE01	NA	mimarks-survey	NA	NA	NA	NA	0	NA	NA	9.777	2011	NA	0	NA	NA	female	NA	NA	NA	drscottkelley@gmail.com	NA	22	NA	".1,g"	NA	NA	"Samples from office surfaces in New York, San Francisco and Tucson"	NA	NA	-74.00597	y	NA	CCME	NA	NA	NA	"People in developed countries spend approximately 90% of their lives indoors, yet we know little about the source and diversity of microbes in built environments. In this study, we combined culture-based cell counting and multiplexed pyrosequencing of environmental ribosomal RNA (rRNA) gene sequences to investigate office space bacterial diversity in three metropolitan areas. Five surfaces common to all offices were sampled using sterile double-tipped swabs, one tip for culturing and one for DNA extraction, in 30 different offices per city (90 offices, 450 total samples). 16S rRNA gene sequences were PCR amplified using bar-coded universal bacterial primers from 54 of the surfaces (18 per city) and pooled for pyrosequencing. A three-factorial Analysis of Variance (ANOVA) found significant differences in viable bacterial abundance between offices inhabited by men or women, among the various surface types, and among cities. Multiplex pyrosequencing identified more than 500 bacterial genera from 20 different bacterial divisions. The most abundant of these genera tended to be common inhabitants of human skin, nasal, oral or intestinal cavities. Other commonly occurring genera appeared to have environmental origins (e.g., soils). There were no significant differences in the bacterial diversity between offices inhabited by men or women or among surfaces, but the bacterial community diversity of the Tucson samples was clearly distinguishable from that of New York and San Francisco, which were indistinguishable. Overall, our comprehensive molecular analysis of office building microbial diversity shows the potential of these methods for studying patterns and origins of indoor bacterial contamination. Humans move through a sea of microbial life that is seldom perceived except in the context of potential disease and decay Feazel et al. (2009)."	ENVO:surface	NA	Office Space Bacterial Abundance and Diversity in Three  Metropolitan Areas	NA	NA	Office Map study	FWD:AGAGTTTGATCCTGGCTCAG; REV:TGCTGCCTCCCGTAGGAGT	22666400	NA	NA	NA	Dan Knights/Justin Kuczinsky	CCME	GAZ:United States of America	NA	NA	0	NA	NA	NA	NA	n	NA	Scott Kelley	Kelley_office_contamination	pyrosequencing	NA	NA	NA	metagenomes	NA	ENVO:terrestrial biome	FLX	NA	NA	NA	NA	NA	Kelley_office_contamination	Kelley_Office_Map	408169	CCME	NA	NA	NA	1479	FLX protocol	NA	NA	n	NA	NA	ENVO:surface	16S rRNA	NA	NA	V2	CCME	NA	Chair	NA	NA	NA	40.71435	NA	NA	Swab_sample_from_woman_chair_in_New_York_office_room_22
NYFC4.522466	TCCACGTCGTCT	CATGCTGCCTCCCGTAGGAGT	NA	NA	NewYork	Chair	NA	NA	Chair	y	2011	Kelley_Office_Map	NA	FWIRNKE01	NA	mimarks-survey	NA	NA	NA	NA	0	NA	NA	9.777	2011	NA	0	NA	NA	female	NA	NA	NA	drscottkelley@gmail.com	NA	4	NA	".1,g"	NA	NA	"Samples from office surfaces in New York, San Francisco and Tucson"	NA	NA	-74.00597	y	NA	CCME	NA	NA	NA	"People in developed countries spend approximately 90% of their lives indoors, yet we know little about the source and diversity of microbes in built environments. In this study, we combined culture-based cell counting and multiplexed pyrosequencing of environmental ribosomal RNA (rRNA) gene sequences to investigate office space bacterial diversity in three metropolitan areas. Five surfaces common to all offices were sampled using sterile double-tipped swabs, one tip for culturing and one for DNA extraction, in 30 different offices per city (90 offices, 450 total samples). 16S rRNA gene sequences were PCR amplified using bar-coded universal bacterial primers from 54 of the surfaces (18 per city) and pooled for pyrosequencing. A three-factorial Analysis of Variance (ANOVA) found significant differences in viable bacterial abundance between offices inhabited by men or women, among the various surface types, and among cities. Multiplex pyrosequencing identified more than 500 bacterial genera from 20 different bacterial divisions. The most abundant of these genera tended to be common inhabitants of human skin, nasal, oral or intestinal cavities. Other commonly occurring genera appeared to have environmental origins (e.g., soils). There were no significant differences in the bacterial diversity between offices inhabited by men or women or among surfaces, but the bacterial community diversity of the Tucson samples was clearly distinguishable from that of New York and San Francisco, which were indistinguishable. Overall, our comprehensive molecular analysis of office building microbial diversity shows the potential of these methods for studying patterns and origins of indoor bacterial contamination. Humans move through a sea of microbial life that is seldom perceived except in the context of potential disease and decay Feazel et al. (2009)."	ENVO:surface	NA	Office Space Bacterial Abundance and Diversity in Three  Metropolitan Areas	NA	NA	Office Map study	FWD:AGAGTTTGATCCTGGCTCAG; REV:TGCTGCCTCCCGTAGGAGT	22666400	NA	NA	NA	Dan Knights/Justin Kuczinsky	CCME	GAZ:United States of America	NA	NA	0	NA	NA	NA	NA	n	NA	Scott Kelley	Kelley_office_contamination	pyrosequencing	NA	NA	NA	metagenomes	NA	ENVO:terrestrial biome	FLX	NA	NA	NA	NA	NA	Kelley_office_contamination	Kelley_Office_Map	408169	CCME	NA	NA	NA	1479	FLX protocol	NA	NA	n	NA	NA	ENVO:surface	16S rRNA	NA	NA	V2	CCME	NA	Chair	NA	NA	NA	40.71435	NA	NA	Swab_sample_from_woman_chair_in_New_York_office_room_4
NYFC7.522459	TCCAGTGCGAGA	CATGCTGCCTCCCGTAGGAGT	NA	NA	NewYork	Chair	NA	NA	Chair	y	2011	Kelley_Office_Map	NA	FWIRNKE01	NA	mimarks-survey	NA	NA	NA	NA	0	NA	NA	9.777	2011	NA	0	NA	NA	female	NA	NA	NA	drscottkelley@gmail.com	NA	7	NA	".1,g"	NA	NA	"Samples from office surfaces in New York, San Francisco and Tucson"	NA	NA	-74.00597	y	NA	CCME	NA	NA	NA	"People in developed countries spend approximately 90% of their lives indoors, yet we know little about the source and diversity of microbes in built environments. In this study, we combined culture-based cell counting and multiplexed pyrosequencing of environmental ribosomal RNA (rRNA) gene sequences to investigate office space bacterial diversity in three metropolitan areas. Five surfaces common to all offices were sampled using sterile double-tipped swabs, one tip for culturing and one for DNA extraction, in 30 different offices per city (90 offices, 450 total samples). 16S rRNA gene sequences were PCR amplified using bar-coded universal bacterial primers from 54 of the surfaces (18 per city) and pooled for pyrosequencing. A three-factorial Analysis of Variance (ANOVA) found significant differences in viable bacterial abundance between offices inhabited by men or women, among the various surface types, and among cities. Multiplex pyrosequencing identified more than 500 bacterial genera from 20 different bacterial divisions. The most abundant of these genera tended to be common inhabitants of human skin, nasal, oral or intestinal cavities. Other commonly occurring genera appeared to have environmental origins (e.g., soils). There were no significant differences in the bacterial diversity between offices inhabited by men or women or among surfaces, but the bacterial community diversity of the Tucson samples was clearly distinguishable from that of New York and San Francisco, which were indistinguishable. Overall, our comprehensive molecular analysis of office building microbial diversity shows the potential of these methods for studying patterns and origins of indoor bacterial contamination. Humans move through a sea of microbial life that is seldom perceived except in the context of potential disease and decay Feazel et al. (2009)."	ENVO:surface	NA	Office Space Bacterial Abundance and Diversity in Three  Metropolitan Areas	NA	NA	Office Map study	FWD:AGAGTTTGATCCTGGCTCAG; REV:TGCTGCCTCCCGTAGGAGT	22666400	NA	NA	NA	Dan Knights/Justin Kuczinsky	CCME	GAZ:United States of America	NA	NA	0	NA	NA	NA	NA	n	NA	Scott Kelley	Kelley_office_contamination	pyrosequencing	NA	NA	NA	metagenomes	NA	ENVO:terrestrial biome	FLX	NA	NA	NA	NA	NA	Kelley_office_contamination	Kelley_Office_Map	408169	CCME	NA	NA	NA	1479	FLX protocol	NA	NA	n	NA	NA	ENVO:surface	16S rRNA	NA	NA	V2	CCME	NA	Chair	NA	NA	NA	40.71435	NA	NA	Swab_sample_from_woman_chair_in_New_York_office_room_7
NYMC18.522437	TCAGCTCAACTA	CATGCTGCCTCCCGTAGGAGT	NA	NA	NewYork	Chair	NA	NA	Chair	y	2011	Kelley_Office_Map	NA	FWIRNKE01	NA	mimarks-survey	NA	NA	NA	NA	0	NA	NA	9.777	2011	NA	0	NA	NA	male	NA	NA	NA	drscottkelley@gmail.com	NA	18	NA	".1,g"	NA	NA	"Samples from office surfaces in New York, San Francisco and Tucson"	NA	NA	-74.00597	y	NA	CCME	NA	NA	NA	"People in developed countries spend approximately 90% of their lives indoors, yet we know little about the source and diversity of microbes in built environments. In this study, we combined culture-based cell counting and multiplexed pyrosequencing of environmental ribosomal RNA (rRNA) gene sequences to investigate office space bacterial diversity in three metropolitan areas. Five surfaces common to all offices were sampled using sterile double-tipped swabs, one tip for culturing and one for DNA extraction, in 30 different offices per city (90 offices, 450 total samples). 16S rRNA gene sequences were PCR amplified using bar-coded universal bacterial primers from 54 of the surfaces (18 per city) and pooled for pyrosequencing. A three-factorial Analysis of Variance (ANOVA) found significant differences in viable bacterial abundance between offices inhabited by men or women, among the various surface types, and among cities. Multiplex pyrosequencing identified more than 500 bacterial genera from 20 different bacterial divisions. The most abundant of these genera tended to be common inhabitants of human skin, nasal, oral or intestinal cavities. Other commonly occurring genera appeared to have environmental origins (e.g., soils). There were no significant differences in the bacterial diversity between offices inhabited by men or women or among surfaces, but the bacterial community diversity of the Tucson samples was clearly distinguishable from that of New York and San Francisco, which were indistinguishable. Overall, our comprehensive molecular analysis of office building microbial diversity shows the potential of these methods for studying patterns and origins of indoor bacterial contamination. Humans move through a sea of microbial life that is seldom perceived except in the context of potential disease and decay Feazel et al. (2009)."	ENVO:surface	NA	Office Space Bacterial Abundance and Diversity in Three  Metropolitan Areas	NA	NA	Office Map study	FWD:AGAGTTTGATCCTGGCTCAG; REV:TGCTGCCTCCCGTAGGAGT	22666400	NA	NA	NA	Dan Knights/Justin Kuczinsky	CCME	GAZ:United States of America	NA	NA	0	NA	NA	NA	NA	n	NA	Scott Kelley	Kelley_office_contamination	pyrosequencing	NA	NA	NA	metagenomes	NA	ENVO:terrestrial biome	FLX	NA	NA	NA	NA	NA	Kelley_office_contamination	Kelley_Office_Map	408169	CCME	NA	NA	NA	1479	FLX protocol	NA	NA	n	NA	NA	ENVO:surface	16S rRNA	NA	NA	V2	CCME	NA	Chair	NA	NA	NA	40.71435	NA	NA	Swab_sample_from_man_chair_in_New_York_office_room_18
NYMC23.522436	TCAGGACTGTGT	CATGCTGCCTCCCGTAGGAGT	NA	NA	NewYork	Chair	NA	NA	Chair	y	2011	Kelley_Office_Map	NA	FWIRNKE01	NA	mimarks-survey	NA	NA	NA	NA	0	NA	NA	9.777	2011	NA	0	NA	NA	male	NA	NA	NA	drscottkelley@gmail.com	NA	23	NA	".1,g"	NA	NA	"Samples from office surfaces in New York, San Francisco and Tucson"	NA	NA	-74.00597	y	NA	CCME	NA	NA	NA	"People in developed countries spend approximately 90% of their lives indoors, yet we know little about the source and diversity of microbes in built environments. In this study, we combined culture-based cell counting and multiplexed pyrosequencing of environmental ribosomal RNA (rRNA) gene sequences to investigate office space bacterial diversity in three metropolitan areas. Five surfaces common to all offices were sampled using sterile double-tipped swabs, one tip for culturing and one for DNA extraction, in 30 different offices per city (90 offices, 450 total samples). 16S rRNA gene sequences were PCR amplified using bar-coded universal bacterial primers from 54 of the surfaces (18 per city) and pooled for pyrosequencing. A three-factorial Analysis of Variance (ANOVA) found significant differences in viable bacterial abundance between offices inhabited by men or women, among the various surface types, and among cities. Multiplex pyrosequencing identified more than 500 bacterial genera from 20 different bacterial divisions. The most abundant of these genera tended to be common inhabitants of human skin, nasal, oral or intestinal cavities. Other commonly occurring genera appeared to have environmental origins (e.g., soils). There were no significant differences in the bacterial diversity between offices inhabited by men or women or among surfaces, but the bacterial community diversity of the Tucson samples was clearly distinguishable from that of New York and San Francisco, which were indistinguishable. Overall, our comprehensive molecular analysis of office building microbial diversity shows the potential of these methods for studying patterns and origins of indoor bacterial contamination. Humans move through a sea of microbial life that is seldom perceived except in the context of potential disease and decay Feazel et al. (2009)."	ENVO:surface	NA	Office Space Bacterial Abundance and Diversity in Three  Metropolitan Areas	NA	NA	Office Map study	FWD:AGAGTTTGATCCTGGCTCAG; REV:TGCTGCCTCCCGTAGGAGT	22666400	NA	NA	NA	Dan Knights/Justin Kuczinsky	CCME	GAZ:United States of America	NA	NA	0	NA	NA	NA	NA	n	NA	Scott Kelley	Kelley_office_contamination	pyrosequencing	NA	NA	NA	metagenomes	NA	ENVO:terrestrial biome	FLX	NA	NA	NA	NA	NA	Kelley_office_contamination	Kelley_Office_Map	408169	CCME	NA	NA	NA	1479	FLX protocol	NA	NA	n	NA	NA	ENVO:surface	16S rRNA	NA	NA	V2	CCME	NA	Chair	NA	NA	NA	40.71435	NA	NA	Swab_sample_from_man_chair_in_New_York_office_room_23
NYMC31.522475	TCAGTACGAGGC	CATGCTGCCTCCCGTAGGAGT	NA	NA	NewYork	Chair	NA	NA	Chair	y	2011	Kelley_Office_Map	NA	FWIRNKE01	NA	mimarks-survey	NA	NA	NA	NA	0	NA	NA	9.777	2011	NA	0	NA	NA	male	NA	NA	NA	drscottkelley@gmail.com	NA	31	NA	".1,g"	NA	NA	"Samples from office surfaces in New York, San Francisco and Tucson"	NA	NA	-74.00597	y	NA	CCME	NA	NA	NA	"People in developed countries spend approximately 90% of their lives indoors, yet we know little about the source and diversity of microbes in built environments. In this study, we combined culture-based cell counting and multiplexed pyrosequencing of environmental ribosomal RNA (rRNA) gene sequences to investigate office space bacterial diversity in three metropolitan areas. Five surfaces common to all offices were sampled using sterile double-tipped swabs, one tip for culturing and one for DNA extraction, in 30 different offices per city (90 offices, 450 total samples). 16S rRNA gene sequences were PCR amplified using bar-coded universal bacterial primers from 54 of the surfaces (18 per city) and pooled for pyrosequencing. A three-factorial Analysis of Variance (ANOVA) found significant differences in viable bacterial abundance between offices inhabited by men or women, among the various surface types, and among cities. Multiplex pyrosequencing identified more than 500 bacterial genera from 20 different bacterial divisions. The most abundant of these genera tended to be common inhabitants of human skin, nasal, oral or intestinal cavities. Other commonly occurring genera appeared to have environmental origins (e.g., soils). There were no significant differences in the bacterial diversity between offices inhabited by men or women or among surfaces, but the bacterial community diversity of the Tucson samples was clearly distinguishable from that of New York and San Francisco, which were indistinguishable. Overall, our comprehensive molecular analysis of office building microbial diversity shows the potential of these methods for studying patterns and origins of indoor bacterial contamination. Humans move through a sea of microbial life that is seldom perceived except in the context of potential disease and decay Feazel et al. (2009)."	ENVO:surface	NA	Office Space Bacterial Abundance and Diversity in Three  Metropolitan Areas	NA	NA	Office Map study	FWD:AGAGTTTGATCCTGGCTCAG; REV:TGCTGCCTCCCGTAGGAGT	22666400	NA	NA	NA	Dan Knights/Justin Kuczinsky	CCME	GAZ:United States of America	NA	NA	0	NA	NA	NA	NA	n	NA	Scott Kelley	Kelley_office_contamination	pyrosequencing	NA	NA	NA	metagenomes	NA	ENVO:terrestrial biome	FLX	NA	NA	NA	NA	NA	Kelley_office_contamination	Kelley_Office_Map	408169	CCME	NA	NA	NA	1479	FLX protocol	NA	NA	n	NA	NA	ENVO:surface	16S rRNA	NA	NA	V2	CCME	NA	Chair	NA	NA	NA	40.71435	NA	NA	Swab_sample_from_man_chair_in_New_York_office_room_31
NYMC5.522451	TCAGCCATGACA	CATGCTGCCTCCCGTAGGAGT	NA	NA	NewYork	Chair	NA	NA	Chair	y	2011	Kelley_Office_Map	NA	FWIRNKE01	NA	mimarks-survey	NA	NA	NA	NA	0	NA	NA	9.777	2011	NA	0	NA	NA	male	NA	NA	NA	drscottkelley@gmail.com	NA	5	NA	".1,g"	NA	NA	"Samples from office surfaces in New York, San Francisco and Tucson"	NA	NA	-74.00597	y	NA	CCME	NA	NA	NA	"People in developed countries spend approximately 90% of their lives indoors, yet we know little about the source and diversity of microbes in built environments. In this study, we combined culture-based cell counting and multiplexed pyrosequencing of environmental ribosomal RNA (rRNA) gene sequences to investigate office space bacterial diversity in three metropolitan areas. Five surfaces common to all offices were sampled using sterile double-tipped swabs, one tip for culturing and one for DNA extraction, in 30 different offices per city (90 offices, 450 total samples). 16S rRNA gene sequences were PCR amplified using bar-coded universal bacterial primers from 54 of the surfaces (18 per city) and pooled for pyrosequencing. A three-factorial Analysis of Variance (ANOVA) found significant differences in viable bacterial abundance between offices inhabited by men or women, among the various surface types, and among cities. Multiplex pyrosequencing identified more than 500 bacterial genera from 20 different bacterial divisions. The most abundant of these genera tended to be common inhabitants of human skin, nasal, oral or intestinal cavities. Other commonly occurring genera appeared to have environmental origins (e.g., soils). There were no significant differences in the bacterial diversity between offices inhabited by men or women or among surfaces, but the bacterial community diversity of the Tucson samples was clearly distinguishable from that of New York and San Francisco, which were indistinguishable. Overall, our comprehensive molecular analysis of office building microbial diversity shows the potential of these methods for studying patterns and origins of indoor bacterial contamination. Humans move through a sea of microbial life that is seldom perceived except in the context of potential disease and decay Feazel et al. (2009)."	ENVO:surface	NA	Office Space Bacterial Abundance and Diversity in Three  Metropolitan Areas	NA	NA	Office Map study	FWD:AGAGTTTGATCCTGGCTCAG; REV:TGCTGCCTCCCGTAGGAGT	22666400	NA	NA	NA	Dan Knights/Justin Kuczinsky	CCME	GAZ:United States of America	NA	NA	0	NA	NA	NA	NA	n	NA	Scott Kelley	Kelley_office_contamination	pyrosequencing	NA	NA	NA	metagenomes	NA	ENVO:terrestrial biome	FLX	NA	NA	NA	NA	NA	Kelley_office_contamination	Kelley_Office_Map	408169	CCME	NA	NA	NA	1479	FLX protocol	NA	NA	n	NA	NA	ENVO:surface	16S rRNA	NA	NA	V2	CCME	NA	Chair	NA	NA	NA	40.71435	NA	NA	Swab_sample_from_man_chair_in_New_York_office_room_5
SFFC10.522446	CGATCGAGTGTT	CATGCTGCCTCCCGTAGGAGT	NA	NA	SanFrancisco	Chair	NA	NA	Chair	y	2011	Kelley_Office_Map	NA	FWIRNKE01	NA	mimarks-survey	NA	NA	NA	NA	0	NA	NA	15.5624	2011	NA	0	NA	NA	female	NA	NA	NA	drscottkelley@gmail.com	NA	10	NA	".1,g"	NA	NA	"Samples from office surfaces in New York, San Francisco and Tucson"	NA	NA	-122.41942	y	NA	CCME	NA	NA	NA	"People in developed countries spend approximately 90% of their lives indoors, yet we know little about the source and diversity of microbes in built environments. In this study, we combined culture-based cell counting and multiplexed pyrosequencing of environmental ribosomal RNA (rRNA) gene sequences to investigate office space bacterial diversity in three metropolitan areas. Five surfaces common to all offices were sampled using sterile double-tipped swabs, one tip for culturing and one for DNA extraction, in 30 different offices per city (90 offices, 450 total samples). 16S rRNA gene sequences were PCR amplified using bar-coded universal bacterial primers from 54 of the surfaces (18 per city) and pooled for pyrosequencing. A three-factorial Analysis of Variance (ANOVA) found significant differences in viable bacterial abundance between offices inhabited by men or women, among the various surface types, and among cities. Multiplex pyrosequencing identified more than 500 bacterial genera from 20 different bacterial divisions. The most abundant of these genera tended to be common inhabitants of human skin, nasal, oral or intestinal cavities. Other commonly occurring genera appeared to have environmental origins (e.g., soils). There were no significant differences in the bacterial diversity between offices inhabited by men or women or among surfaces, but the bacterial community diversity of the Tucson samples was clearly distinguishable from that of New York and San Francisco, which were indistinguishable. Overall, our comprehensive molecular analysis of office building microbial diversity shows the potential of these methods for studying patterns and origins of indoor bacterial contamination. Humans move through a sea of microbial life that is seldom perceived except in the context of potential disease and decay Feazel et al. (2009)."	ENVO:surface	NA	Office Space Bacterial Abundance and Diversity in Three  Metropolitan Areas	NA	NA	Office Map study	FWD:AGAGTTTGATCCTGGCTCAG; REV:TGCTGCCTCCCGTAGGAGT	22666400	NA	NA	NA	Dan Knights/Justin Kuczinsky	CCME	GAZ:United States of America	NA	NA	0	NA	NA	NA	NA	n	NA	Scott Kelley	Kelley_office_contamination	pyrosequencing	NA	NA	NA	metagenomes	NA	ENVO:terrestrial biome	FLX	NA	NA	NA	NA	NA	Kelley_office_contamination	Kelley_Office_Map	408169	CCME	NA	NA	NA	1479	FLX protocol	NA	NA	n	NA	NA	ENVO:surface	16S rRNA	NA	NA	V2	CCME	NA	Chair	NA	NA	NA	37.77493	NA	NA	Swab_sample_from_woman_chair_in_San_Francisco_office_room_10
SFFC2.522476	CGATAGATCTTC	CATGCTGCCTCCCGTAGGAGT	NA	NA	SanFrancisco	Chair	NA	NA	Chair	y	2011	Kelley_Office_Map	NA	FWIRNKE01	NA	mimarks-survey	NA	NA	NA	NA	0	NA	NA	15.5624	2011	NA	0	NA	NA	female	NA	NA	NA	drscottkelley@gmail.com	NA	2	NA	".1,g"	NA	NA	"Samples from office surfaces in New York, San Francisco and Tucson"	NA	NA	-122.41942	y	NA	CCME	NA	NA	NA	"People in developed countries spend approximately 90% of their lives indoors, yet we know little about the source and diversity of microbes in built environments. In this study, we combined culture-based cell counting and multiplexed pyrosequencing of environmental ribosomal RNA (rRNA) gene sequences to investigate office space bacterial diversity in three metropolitan areas. Five surfaces common to all offices were sampled using sterile double-tipped swabs, one tip for culturing and one for DNA extraction, in 30 different offices per city (90 offices, 450 total samples). 16S rRNA gene sequences were PCR amplified using bar-coded universal bacterial primers from 54 of the surfaces (18 per city) and pooled for pyrosequencing. A three-factorial Analysis of Variance (ANOVA) found significant differences in viable bacterial abundance between offices inhabited by men or women, among the various surface types, and among cities. Multiplex pyrosequencing identified more than 500 bacterial genera from 20 different bacterial divisions. The most abundant of these genera tended to be common inhabitants of human skin, nasal, oral or intestinal cavities. Other commonly occurring genera appeared to have environmental origins (e.g., soils). There were no significant differences in the bacterial diversity between offices inhabited by men or women or among surfaces, but the bacterial community diversity of the Tucson samples was clearly distinguishable from that of New York and San Francisco, which were indistinguishable. Overall, our comprehensive molecular analysis of office building microbial diversity shows the potential of these methods for studying patterns and origins of indoor bacterial contamination. Humans move through a sea of microbial life that is seldom perceived except in the context of potential disease and decay Feazel et al. (2009)."	ENVO:surface	NA	Office Space Bacterial Abundance and Diversity in Three  Metropolitan Areas	NA	NA	Office Map study	FWD:AGAGTTTGATCCTGGCTCAG; REV:TGCTGCCTCCCGTAGGAGT	22666400	NA	NA	NA	Dan Knights/Justin Kuczinsky	CCME	GAZ:United States of America	NA	NA	0	NA	NA	NA	NA	n	NA	Scott Kelley	Kelley_office_contamination	pyrosequencing	NA	NA	NA	metagenomes	NA	ENVO:terrestrial biome	FLX	NA	NA	NA	NA	NA	Kelley_office_contamination	Kelley_Office_Map	408169	CCME	NA	NA	NA	1479	FLX protocol	NA	NA	n	NA	NA	ENVO:surface	16S rRNA	NA	NA	V2	CCME	NA	Chair	NA	NA	NA	37.77493	NA	NA	Swab_sample_from_woman_chair_in_San_Francisco_office_room_2
SFFC23.522471	CGATGCACCAGA	CATGCTGCCTCCCGTAGGAGT	NA	NA	SanFrancisco	Chair	NA	NA	Chair	y	2011	Kelley_Office_Map	NA	FWIRNKE01	NA	mimarks-survey	NA	NA	NA	NA	0	NA	NA	15.5624	2011	NA	0	NA	NA	female	NA	NA	NA	drscottkelley@gmail.com	NA	23	NA	".1,g"	NA	NA	"Samples from office surfaces in New York, San Francisco and Tucson"	NA	NA	-122.41942	y	NA	CCME	NA	NA	NA	"People in developed countries spend approximately 90% of their lives indoors, yet we know little about the source and diversity of microbes in built environments. In this study, we combined culture-based cell counting and multiplexed pyrosequencing of environmental ribosomal RNA (rRNA) gene sequences to investigate office space bacterial diversity in three metropolitan areas. Five surfaces common to all offices were sampled using sterile double-tipped swabs, one tip for culturing and one for DNA extraction, in 30 different offices per city (90 offices, 450 total samples). 16S rRNA gene sequences were PCR amplified using bar-coded universal bacterial primers from 54 of the surfaces (18 per city) and pooled for pyrosequencing. A three-factorial Analysis of Variance (ANOVA) found significant differences in viable bacterial abundance between offices inhabited by men or women, among the various surface types, and among cities. Multiplex pyrosequencing identified more than 500 bacterial genera from 20 different bacterial divisions. The most abundant of these genera tended to be common inhabitants of human skin, nasal, oral or intestinal cavities. Other commonly occurring genera appeared to have environmental origins (e.g., soils). There were no significant differences in the bacterial diversity between offices inhabited by men or women or among surfaces, but the bacterial community diversity of the Tucson samples was clearly distinguishable from that of New York and San Francisco, which were indistinguishable. Overall, our comprehensive molecular analysis of office building microbial diversity shows the potential of these methods for studying patterns and origins of indoor bacterial contamination. Humans move through a sea of microbial life that is seldom perceived except in the context of potential disease and decay Feazel et al. (2009)."	ENVO:surface	NA	Office Space Bacterial Abundance and Diversity in Three  Metropolitan Areas	NA	NA	Office Map study	FWD:AGAGTTTGATCCTGGCTCAG; REV:TGCTGCCTCCCGTAGGAGT	22666400	NA	NA	NA	Dan Knights/Justin Kuczinsky	CCME	GAZ:United States of America	NA	NA	0	NA	NA	NA	NA	n	NA	Scott Kelley	Kelley_office_contamination	pyrosequencing	NA	NA	NA	metagenomes	NA	ENVO:terrestrial biome	FLX	NA	NA	NA	NA	NA	Kelley_office_contamination	Kelley_Office_Map	408169	CCME	NA	NA	NA	1479	FLX protocol	NA	NA	n	NA	NA	ENVO:surface	16S rRNA	NA	NA	V2	CCME	NA	Chair	NA	NA	NA	37.77493	NA	NA	Swab_sample_from_woman_chair_in_San_Francisco_office_room_23
SFFC24.522467	CGATGTCGTCAA	CATGCTGCCTCCCGTAGGAGT	NA	NA	SanFrancisco	Chair	NA	NA	Chair	y	2011	Kelley_Office_Map	NA	FWIRNKE01	NA	mimarks-survey	NA	NA	NA	NA	0	NA	NA	15.5624	2011	NA	0	NA	NA	female	NA	NA	NA	drscottkelley@gmail.com	NA	24	NA	".1,g"	NA	NA	"Samples from office surfaces in New York, San Francisco and Tucson"	NA	NA	-122.41942	y	NA	CCME	NA	NA	NA	"People in developed countries spend approximately 90% of their lives indoors, yet we know little about the source and diversity of microbes in built environments. In this study, we combined culture-based cell counting and multiplexed pyrosequencing of environmental ribosomal RNA (rRNA) gene sequences to investigate office space bacterial diversity in three metropolitan areas. Five surfaces common to all offices were sampled using sterile double-tipped swabs, one tip for culturing and one for DNA extraction, in 30 different offices per city (90 offices, 450 total samples). 16S rRNA gene sequences were PCR amplified using bar-coded universal bacterial primers from 54 of the surfaces (18 per city) and pooled for pyrosequencing. A three-factorial Analysis of Variance (ANOVA) found significant differences in viable bacterial abundance between offices inhabited by men or women, among the various surface types, and among cities. Multiplex pyrosequencing identified more than 500 bacterial genera from 20 different bacterial divisions. The most abundant of these genera tended to be common inhabitants of human skin, nasal, oral or intestinal cavities. Other commonly occurring genera appeared to have environmental origins (e.g., soils). There were no significant differences in the bacterial diversity between offices inhabited by men or women or among surfaces, but the bacterial community diversity of the Tucson samples was clearly distinguishable from that of New York and San Francisco, which were indistinguishable. Overall, our comprehensive molecular analysis of office building microbial diversity shows the potential of these methods for studying patterns and origins of indoor bacterial contamination. Humans move through a sea of microbial life that is seldom perceived except in the context of potential disease and decay Feazel et al. (2009)."	ENVO:surface	NA	Office Space Bacterial Abundance and Diversity in Three  Metropolitan Areas	NA	NA	Office Map study	FWD:AGAGTTTGATCCTGGCTCAG; REV:TGCTGCCTCCCGTAGGAGT	22666400	NA	NA	NA	Dan Knights/Justin Kuczinsky	CCME	GAZ:United States of America	NA	NA	0	NA	NA	NA	NA	n	NA	Scott Kelley	Kelley_office_contamination	pyrosequencing	NA	NA	NA	metagenomes	NA	ENVO:terrestrial biome	FLX	NA	NA	NA	NA	NA	Kelley_office_contamination	Kelley_Office_Map	408169	CCME	NA	NA	NA	1479	FLX protocol	NA	NA	n	NA	NA	ENVO:surface	16S rRNA	NA	NA	V2	CCME	NA	Chair	NA	NA	NA	37.77493	NA	NA	Swab_sample_from_woman_chair_in_San_Francisco_office_room_24
SFFC3.522434	CGATATTCATCG	CATGCTGCCTCCCGTAGGAGT	NA	NA	SanFrancisco	Chair	NA	NA	Chair	y	2011	Kelley_Office_Map	NA	FWIRNKE01	NA	mimarks-survey	NA	NA	NA	NA	0	NA	NA	15.5624	2011	NA	0	NA	NA	female	NA	NA	NA	drscottkelley@gmail.com	NA	3	NA	".1,g"	NA	NA	"Samples from office surfaces in New York, San Francisco and Tucson"	NA	NA	-122.41942	y	NA	CCME	NA	NA	NA	"People in developed countries spend approximately 90% of their lives indoors, yet we know little about the source and diversity of microbes in built environments. In this study, we combined culture-based cell counting and multiplexed pyrosequencing of environmental ribosomal RNA (rRNA) gene sequences to investigate office space bacterial diversity in three metropolitan areas. Five surfaces common to all offices were sampled using sterile double-tipped swabs, one tip for culturing and one for DNA extraction, in 30 different offices per city (90 offices, 450 total samples). 16S rRNA gene sequences were PCR amplified using bar-coded universal bacterial primers from 54 of the surfaces (18 per city) and pooled for pyrosequencing. A three-factorial Analysis of Variance (ANOVA) found significant differences in viable bacterial abundance between offices inhabited by men or women, among the various surface types, and among cities. Multiplex pyrosequencing identified more than 500 bacterial genera from 20 different bacterial divisions. The most abundant of these genera tended to be common inhabitants of human skin, nasal, oral or intestinal cavities. Other commonly occurring genera appeared to have environmental origins (e.g., soils). There were no significant differences in the bacterial diversity between offices inhabited by men or women or among surfaces, but the bacterial community diversity of the Tucson samples was clearly distinguishable from that of New York and San Francisco, which were indistinguishable. Overall, our comprehensive molecular analysis of office building microbial diversity shows the potential of these methods for studying patterns and origins of indoor bacterial contamination. Humans move through a sea of microbial life that is seldom perceived except in the context of potential disease and decay Feazel et al. (2009)."	ENVO:surface	NA	Office Space Bacterial Abundance and Diversity in Three  Metropolitan Areas	NA	NA	Office Map study	FWD:AGAGTTTGATCCTGGCTCAG; REV:TGCTGCCTCCCGTAGGAGT	22666400	NA	NA	NA	Dan Knights/Justin Kuczinsky	CCME	GAZ:United States of America	NA	NA	0	NA	NA	NA	NA	n	NA	Scott Kelley	Kelley_office_contamination	pyrosequencing	NA	NA	NA	metagenomes	NA	ENVO:terrestrial biome	FLX	NA	NA	NA	NA	NA	Kelley_office_contamination	Kelley_Office_Map	408169	CCME	NA	NA	NA	1479	FLX protocol	NA	NA	n	NA	NA	ENVO:surface	16S rRNA	NA	NA	V2	CCME	NA	Chair	NA	NA	NA	37.77493	NA	NA	Swab_sample_from_woman_chair_in_San_Francisco_office_room_3
SFMC16.522482	CGACAGCTGACA	CATGCTGCCTCCCGTAGGAGT	NA	NA	SanFrancisco	Chair	NA	NA	Chair	y	2011	Kelley_Office_Map	NA	FWIRNKE01	NA	mimarks-survey	NA	NA	NA	NA	0	NA	NA	15.5624	2011	NA	0	NA	NA	male	NA	NA	NA	drscottkelley@gmail.com	NA	16	NA	".1,g"	NA	NA	"Samples from office surfaces in New York, San Francisco and Tucson"	NA	NA	-122.41942	y	NA	CCME	NA	NA	NA	"People in developed countries spend approximately 90% of their lives indoors, yet we know little about the source and diversity of microbes in built environments. In this study, we combined culture-based cell counting and multiplexed pyrosequencing of environmental ribosomal RNA (rRNA) gene sequences to investigate office space bacterial diversity in three metropolitan areas. Five surfaces common to all offices were sampled using sterile double-tipped swabs, one tip for culturing and one for DNA extraction, in 30 different offices per city (90 offices, 450 total samples). 16S rRNA gene sequences were PCR amplified using bar-coded universal bacterial primers from 54 of the surfaces (18 per city) and pooled for pyrosequencing. A three-factorial Analysis of Variance (ANOVA) found significant differences in viable bacterial abundance between offices inhabited by men or women, among the various surface types, and among cities. Multiplex pyrosequencing identified more than 500 bacterial genera from 20 different bacterial divisions. The most abundant of these genera tended to be common inhabitants of human skin, nasal, oral or intestinal cavities. Other commonly occurring genera appeared to have environmental origins (e.g., soils). There were no significant differences in the bacterial diversity between offices inhabited by men or women or among surfaces, but the bacterial community diversity of the Tucson samples was clearly distinguishable from that of New York and San Francisco, which were indistinguishable. Overall, our comprehensive molecular analysis of office building microbial diversity shows the potential of these methods for studying patterns and origins of indoor bacterial contamination. Humans move through a sea of microbial life that is seldom perceived except in the context of potential disease and decay Feazel et al. (2009)."	ENVO:surface	NA	Office Space Bacterial Abundance and Diversity in Three  Metropolitan Areas	NA	NA	Office Map study	FWD:AGAGTTTGATCCTGGCTCAG; REV:TGCTGCCTCCCGTAGGAGT	22666400	NA	NA	NA	Dan Knights/Justin Kuczinsky	CCME	GAZ:United States of America	NA	NA	0	NA	NA	NA	NA	n	NA	Scott Kelley	Kelley_office_contamination	pyrosequencing	NA	NA	NA	metagenomes	NA	ENVO:terrestrial biome	FLX	NA	NA	NA	NA	NA	Kelley_office_contamination	Kelley_Office_Map	408169	CCME	NA	NA	NA	1479	FLX protocol	NA	NA	n	NA	NA	ENVO:surface	16S rRNA	NA	NA	V2	CCME	NA	Chair	NA	NA	NA	37.77493	NA	NA	Swab_sample_from_man_chair_in_San_Francisco_office_room_16
SFMC28.522435	CGACATGCTATT	CATGCTGCCTCCCGTAGGAGT	NA	NA	SanFrancisco	Chair	NA	NA	Chair	y	2011	Kelley_Office_Map	NA	FWIRNKE01	NA	mimarks-survey	NA	NA	NA	NA	0	NA	NA	15.5624	2011	NA	0	NA	NA	male	NA	NA	NA	drscottkelley@gmail.com	NA	28	NA	".1,g"	NA	NA	"Samples from office surfaces in New York, San Francisco and Tucson"	NA	NA	-122.41942	y	NA	CCME	NA	NA	NA	"People in developed countries spend approximately 90% of their lives indoors, yet we know little about the source and diversity of microbes in built environments. In this study, we combined culture-based cell counting and multiplexed pyrosequencing of environmental ribosomal RNA (rRNA) gene sequences to investigate office space bacterial diversity in three metropolitan areas. Five surfaces common to all offices were sampled using sterile double-tipped swabs, one tip for culturing and one for DNA extraction, in 30 different offices per city (90 offices, 450 total samples). 16S rRNA gene sequences were PCR amplified using bar-coded universal bacterial primers from 54 of the surfaces (18 per city) and pooled for pyrosequencing. A three-factorial Analysis of Variance (ANOVA) found significant differences in viable bacterial abundance between offices inhabited by men or women, among the various surface types, and among cities. Multiplex pyrosequencing identified more than 500 bacterial genera from 20 different bacterial divisions. The most abundant of these genera tended to be common inhabitants of human skin, nasal, oral or intestinal cavities. Other commonly occurring genera appeared to have environmental origins (e.g., soils). There were no significant differences in the bacterial diversity between offices inhabited by men or women or among surfaces, but the bacterial community diversity of the Tucson samples was clearly distinguishable from that of New York and San Francisco, which were indistinguishable. Overall, our comprehensive molecular analysis of office building microbial diversity shows the potential of these methods for studying patterns and origins of indoor bacterial contamination. Humans move through a sea of microbial life that is seldom perceived except in the context of potential disease and decay Feazel et al. (2009)."	ENVO:surface	NA	Office Space Bacterial Abundance and Diversity in Three  Metropolitan Areas	NA	NA	Office Map study	FWD:AGAGTTTGATCCTGGCTCAG; REV:TGCTGCCTCCCGTAGGAGT	22666400	NA	NA	NA	Dan Knights/Justin Kuczinsky	CCME	GAZ:United States of America	NA	NA	0	NA	NA	NA	NA	n	NA	Scott Kelley	Kelley_office_contamination	pyrosequencing	NA	NA	NA	metagenomes	NA	ENVO:terrestrial biome	FLX	NA	NA	NA	NA	NA	Kelley_office_contamination	Kelley_Office_Map	408169	CCME	NA	NA	NA	1479	FLX protocol	NA	NA	n	NA	NA	ENVO:surface	16S rRNA	NA	NA	V2	CCME	NA	Chair	NA	NA	NA	37.77493	NA	NA	Swab_sample_from_man_chair_in_San_Francisco_office_room_28
SFMC29.522447	CGACTTATGTGT	CATGCTGCCTCCCGTAGGAGT	NA	NA	SanFrancisco	Chair	NA	NA	Chair	y	2011	Kelley_Office_Map	NA	FWIRNKE01	NA	mimarks-survey	NA	NA	NA	NA	0	NA	NA	15.5624	2011	NA	0	NA	NA	male	NA	NA	NA	drscottkelley@gmail.com	NA	29	NA	".1,g"	NA	NA	"Samples from office surfaces in New York, San Francisco and Tucson"	NA	NA	-122.41942	y	NA	CCME	NA	NA	NA	"People in developed countries spend approximately 90% of their lives indoors, yet we know little about the source and diversity of microbes in built environments. In this study, we combined culture-based cell counting and multiplexed pyrosequencing of environmental ribosomal RNA (rRNA) gene sequences to investigate office space bacterial diversity in three metropolitan areas. Five surfaces common to all offices were sampled using sterile double-tipped swabs, one tip for culturing and one for DNA extraction, in 30 different offices per city (90 offices, 450 total samples). 16S rRNA gene sequences were PCR amplified using bar-coded universal bacterial primers from 54 of the surfaces (18 per city) and pooled for pyrosequencing. A three-factorial Analysis of Variance (ANOVA) found significant differences in viable bacterial abundance between offices inhabited by men or women, among the various surface types, and among cities. Multiplex pyrosequencing identified more than 500 bacterial genera from 20 different bacterial divisions. The most abundant of these genera tended to be common inhabitants of human skin, nasal, oral or intestinal cavities. Other commonly occurring genera appeared to have environmental origins (e.g., soils). There were no significant differences in the bacterial diversity between offices inhabited by men or women or among surfaces, but the bacterial community diversity of the Tucson samples was clearly distinguishable from that of New York and San Francisco, which were indistinguishable. Overall, our comprehensive molecular analysis of office building microbial diversity shows the potential of these methods for studying patterns and origins of indoor bacterial contamination. Humans move through a sea of microbial life that is seldom perceived except in the context of potential disease and decay Feazel et al. (2009)."	ENVO:surface	NA	Office Space Bacterial Abundance and Diversity in Three  Metropolitan Areas	NA	NA	Office Map study	FWD:AGAGTTTGATCCTGGCTCAG; REV:TGCTGCCTCCCGTAGGAGT	22666400	NA	NA	NA	Dan Knights/Justin Kuczinsky	CCME	GAZ:United States of America	NA	NA	0	NA	NA	NA	NA	n	NA	Scott Kelley	Kelley_office_contamination	pyrosequencing	NA	NA	NA	metagenomes	NA	ENVO:terrestrial biome	FLX	NA	NA	NA	NA	NA	Kelley_office_contamination	Kelley_Office_Map	408169	CCME	NA	NA	NA	1479	FLX protocol	NA	NA	n	NA	NA	ENVO:surface	16S rRNA	NA	NA	V2	CCME	NA	Chair	NA	NA	NA	37.77493	NA	NA	Swab_sample_from_man_chair_in_San_Francisco_office_room_29
SFMC5.522444	CGAAGACTGCTG	CATGCTGCCTCCCGTAGGAGT	NA	NA	SanFrancisco	Chair	NA	NA	Chair	y	2011	Kelley_Office_Map	NA	FWIRNKE01	NA	mimarks-survey	NA	NA	NA	NA	0	NA	NA	15.5624	2011	NA	0	NA	NA	male	NA	NA	NA	drscottkelley@gmail.com	NA	5	NA	".1,g"	NA	NA	"Samples from office surfaces in New York, San Francisco and Tucson"	NA	NA	-122.41942	y	NA	CCME	NA	NA	NA	"People in developed countries spend approximately 90% of their lives indoors, yet we know little about the source and diversity of microbes in built environments. In this study, we combined culture-based cell counting and multiplexed pyrosequencing of environmental ribosomal RNA (rRNA) gene sequences to investigate office space bacterial diversity in three metropolitan areas. Five surfaces common to all offices were sampled using sterile double-tipped swabs, one tip for culturing and one for DNA extraction, in 30 different offices per city (90 offices, 450 total samples). 16S rRNA gene sequences were PCR amplified using bar-coded universal bacterial primers from 54 of the surfaces (18 per city) and pooled for pyrosequencing. A three-factorial Analysis of Variance (ANOVA) found significant differences in viable bacterial abundance between offices inhabited by men or women, among the various surface types, and among cities. Multiplex pyrosequencing identified more than 500 bacterial genera from 20 different bacterial divisions. The most abundant of these genera tended to be common inhabitants of human skin, nasal, oral or intestinal cavities. Other commonly occurring genera appeared to have environmental origins (e.g., soils). There were no significant differences in the bacterial diversity between offices inhabited by men or women or among surfaces, but the bacterial community diversity of the Tucson samples was clearly distinguishable from that of New York and San Francisco, which were indistinguishable. Overall, our comprehensive molecular analysis of office building microbial diversity shows the potential of these methods for studying patterns and origins of indoor bacterial contamination. Humans move through a sea of microbial life that is seldom perceived except in the context of potential disease and decay Feazel et al. (2009)."	ENVO:surface	NA	Office Space Bacterial Abundance and Diversity in Three  Metropolitan Areas	NA	NA	Office Map study	FWD:AGAGTTTGATCCTGGCTCAG; REV:TGCTGCCTCCCGTAGGAGT	22666400	NA	NA	NA	Dan Knights/Justin Kuczinsky	CCME	GAZ:United States of America	NA	NA	0	NA	NA	NA	NA	n	NA	Scott Kelley	Kelley_office_contamination	pyrosequencing	NA	NA	NA	metagenomes	NA	ENVO:terrestrial biome	FLX	NA	NA	NA	NA	NA	Kelley_office_contamination	Kelley_Office_Map	408169	CCME	NA	NA	NA	1479	FLX protocol	NA	NA	n	NA	NA	ENVO:surface	16S rRNA	NA	NA	V2	CCME	NA	Chair	NA	NA	NA	37.77493	NA	NA	Swab_sample_from_man_chair_in_San_Francisco_office_room_5
SFMC7.522456	CGAATCGACACT	CATGCTGCCTCCCGTAGGAGT	NA	NA	SanFrancisco	Chair	NA	NA	Chair	y	2011	Kelley_Office_Map	NA	FWIRNKE01	NA	mimarks-survey	NA	NA	NA	NA	0	NA	NA	15.5624	2011	NA	0	NA	NA	male	NA	NA	NA	drscottkelley@gmail.com	NA	7	NA	".1,g"	NA	NA	"Samples from office surfaces in New York, San Francisco and Tucson"	NA	NA	-122.41942	y	NA	CCME	NA	NA	NA	"People in developed countries spend approximately 90% of their lives indoors, yet we know little about the source and diversity of microbes in built environments. In this study, we combined culture-based cell counting and multiplexed pyrosequencing of environmental ribosomal RNA (rRNA) gene sequences to investigate office space bacterial diversity in three metropolitan areas. Five surfaces common to all offices were sampled using sterile double-tipped swabs, one tip for culturing and one for DNA extraction, in 30 different offices per city (90 offices, 450 total samples). 16S rRNA gene sequences were PCR amplified using bar-coded universal bacterial primers from 54 of the surfaces (18 per city) and pooled for pyrosequencing. A three-factorial Analysis of Variance (ANOVA) found significant differences in viable bacterial abundance between offices inhabited by men or women, among the various surface types, and among cities. Multiplex pyrosequencing identified more than 500 bacterial genera from 20 different bacterial divisions. The most abundant of these genera tended to be common inhabitants of human skin, nasal, oral or intestinal cavities. Other commonly occurring genera appeared to have environmental origins (e.g., soils). There were no significant differences in the bacterial diversity between offices inhabited by men or women or among surfaces, but the bacterial community diversity of the Tucson samples was clearly distinguishable from that of New York and San Francisco, which were indistinguishable. Overall, our comprehensive molecular analysis of office building microbial diversity shows the potential of these methods for studying patterns and origins of indoor bacterial contamination. Humans move through a sea of microbial life that is seldom perceived except in the context of potential disease and decay Feazel et al. (2009)."	ENVO:surface	NA	Office Space Bacterial Abundance and Diversity in Three  Metropolitan Areas	NA	NA	Office Map study	FWD:AGAGTTTGATCCTGGCTCAG; REV:TGCTGCCTCCCGTAGGAGT	22666400	NA	NA	NA	Dan Knights/Justin Kuczinsky	CCME	GAZ:United States of America	NA	NA	0	NA	NA	NA	NA	n	NA	Scott Kelley	Kelley_office_contamination	pyrosequencing	NA	NA	NA	metagenomes	NA	ENVO:terrestrial biome	FLX	NA	NA	NA	NA	NA	Kelley_office_contamination	Kelley_Office_Map	408169	CCME	NA	NA	NA	1479	FLX protocol	NA	NA	n	NA	NA	ENVO:surface	16S rRNA	NA	NA	V2	CCME	NA	Chair	NA	NA	NA	37.77493	NA	NA	Swab_sample_from_man_chair_in_San_Francisco_office_room_7
TUFC13.522473	CGTAGAACGTGC	CATGCTGCCTCCCGTAGGAGT	NA	NA	Tucson	Chair	NA	NA	Chair	y	2011	Kelley_Office_Map	NA	FWIRNKE01	NA	mimarks-survey	NA	NA	NA	NA	0	NA	NA	758.65747	2011	NA	0	NA	NA	female	NA	NA	NA	drscottkelley@gmail.com	NA	13	NA	".1,g"	NA	NA	"Samples from office surfaces in New York, San Francisco and Tucson"	NA	NA	-110.9264	y	NA	CCME	NA	NA	NA	"People in developed countries spend approximately 90% of their lives indoors, yet we know little about the source and diversity of microbes in built environments. In this study, we combined culture-based cell counting and multiplexed pyrosequencing of environmental ribosomal RNA (rRNA) gene sequences to investigate office space bacterial diversity in three metropolitan areas. Five surfaces common to all offices were sampled using sterile double-tipped swabs, one tip for culturing and one for DNA extraction, in 30 different offices per city (90 offices, 450 total samples). 16S rRNA gene sequences were PCR amplified using bar-coded universal bacterial primers from 54 of the surfaces (18 per city) and pooled for pyrosequencing. A three-factorial Analysis of Variance (ANOVA) found significant differences in viable bacterial abundance between offices inhabited by men or women, among the various surface types, and among cities. Multiplex pyrosequencing identified more than 500 bacterial genera from 20 different bacterial divisions. The most abundant of these genera tended to be common inhabitants of human skin, nasal, oral or intestinal cavities. Other commonly occurring genera appeared to have environmental origins (e.g., soils). There were no significant differences in the bacterial diversity between offices inhabited by men or women or among surfaces, but the bacterial community diversity of the Tucson samples was clearly distinguishable from that of New York and San Francisco, which were indistinguishable. Overall, our comprehensive molecular analysis of office building microbial diversity shows the potential of these methods for studying patterns and origins of indoor bacterial contamination. Humans move through a sea of microbial life that is seldom perceived except in the context of potential disease and decay Feazel et al. (2009)."	ENVO:surface	NA	Office Space Bacterial Abundance and Diversity in Three  Metropolitan Areas	NA	NA	Office Map study	FWD:AGAGTTTGATCCTGGCTCAG; REV:TGCTGCCTCCCGTAGGAGT	22666400	NA	NA	NA	Dan Knights/Justin Kuczinsky	CCME	GAZ:United States of America	NA	NA	0	NA	NA	NA	NA	n	NA	Scott Kelley	Kelley_office_contamination	pyrosequencing	NA	NA	NA	metagenomes	NA	ENVO:terrestrial biome	FLX	NA	NA	NA	NA	NA	Kelley_office_contamination	Kelley_Office_Map	408169	CCME	NA	NA	NA	1479	FLX protocol	NA	NA	n	NA	NA	ENVO:surface	16S rRNA	NA	NA	V2	CCME	NA	Chair	NA	NA	NA	32.22174	NA	NA	Swab_sample_from_woman_chair_in_Tucson_office_room_13
TUFC19.522472	CGTATCTGCGAA	CATGCTGCCTCCCGTAGGAGT	NA	NA	Tucson	Chair	NA	NA	Chair	y	2011	Kelley_Office_Map	NA	FWIRNKE01	NA	mimarks-survey	NA	NA	NA	NA	0	NA	NA	758.65747	2011	NA	0	NA	NA	female	NA	NA	NA	drscottkelley@gmail.com	NA	19	NA	".1,g"	NA	NA	"Samples from office surfaces in New York, San Francisco and Tucson"	NA	NA	-110.9264	y	NA	CCME	NA	NA	NA	"People in developed countries spend approximately 90% of their lives indoors, yet we know little about the source and diversity of microbes in built environments. In this study, we combined culture-based cell counting and multiplexed pyrosequencing of environmental ribosomal RNA (rRNA) gene sequences to investigate office space bacterial diversity in three metropolitan areas. Five surfaces common to all offices were sampled using sterile double-tipped swabs, one tip for culturing and one for DNA extraction, in 30 different offices per city (90 offices, 450 total samples). 16S rRNA gene sequences were PCR amplified using bar-coded universal bacterial primers from 54 of the surfaces (18 per city) and pooled for pyrosequencing. A three-factorial Analysis of Variance (ANOVA) found significant differences in viable bacterial abundance between offices inhabited by men or women, among the various surface types, and among cities. Multiplex pyrosequencing identified more than 500 bacterial genera from 20 different bacterial divisions. The most abundant of these genera tended to be common inhabitants of human skin, nasal, oral or intestinal cavities. Other commonly occurring genera appeared to have environmental origins (e.g., soils). There were no significant differences in the bacterial diversity between offices inhabited by men or women or among surfaces, but the bacterial community diversity of the Tucson samples was clearly distinguishable from that of New York and San Francisco, which were indistinguishable. Overall, our comprehensive molecular analysis of office building microbial diversity shows the potential of these methods for studying patterns and origins of indoor bacterial contamination. Humans move through a sea of microbial life that is seldom perceived except in the context of potential disease and decay Feazel et al. (2009)."	ENVO:surface	NA	Office Space Bacterial Abundance and Diversity in Three  Metropolitan Areas	NA	NA	Office Map study	FWD:AGAGTTTGATCCTGGCTCAG; REV:TGCTGCCTCCCGTAGGAGT	22666400	NA	NA	NA	Dan Knights/Justin Kuczinsky	CCME	GAZ:United States of America	NA	NA	0	NA	NA	NA	NA	n	NA	Scott Kelley	Kelley_office_contamination	pyrosequencing	NA	NA	NA	metagenomes	NA	ENVO:terrestrial biome	FLX	NA	NA	NA	NA	NA	Kelley_office_contamination	Kelley_Office_Map	408169	CCME	NA	NA	NA	1479	FLX protocol	NA	NA	n	NA	NA	ENVO:surface	16S rRNA	NA	NA	V2	CCME	NA	Chair	NA	NA	NA	32.22174	NA	NA	Swab_sample_from_woman_chair_in_Tucson_office_room_19
TUFC22.522433	CGTATGCTGTAT	CATGCTGCCTCCCGTAGGAGT	NA	NA	Tucson	Chair	NA	NA	Chair	y	2011	Kelley_Office_Map	NA	FWIRNKE01	NA	mimarks-survey	NA	NA	NA	NA	0	NA	NA	758.65747	2011	NA	0	NA	NA	female	NA	NA	NA	drscottkelley@gmail.com	NA	22	NA	".1,g"	NA	NA	"Samples from office surfaces in New York, San Francisco and Tucson"	NA	NA	-110.9264	y	NA	CCME	NA	NA	NA	"People in developed countries spend approximately 90% of their lives indoors, yet we know little about the source and diversity of microbes in built environments. In this study, we combined culture-based cell counting and multiplexed pyrosequencing of environmental ribosomal RNA (rRNA) gene sequences to investigate office space bacterial diversity in three metropolitan areas. Five surfaces common to all offices were sampled using sterile double-tipped swabs, one tip for culturing and one for DNA extraction, in 30 different offices per city (90 offices, 450 total samples). 16S rRNA gene sequences were PCR amplified using bar-coded universal bacterial primers from 54 of the surfaces (18 per city) and pooled for pyrosequencing. A three-factorial Analysis of Variance (ANOVA) found significant differences in viable bacterial abundance between offices inhabited by men or women, among the various surface types, and among cities. Multiplex pyrosequencing identified more than 500 bacterial genera from 20 different bacterial divisions. The most abundant of these genera tended to be common inhabitants of human skin, nasal, oral or intestinal cavities. Other commonly occurring genera appeared to have environmental origins (e.g., soils). There were no significant differences in the bacterial diversity between offices inhabited by men or women or among surfaces, but the bacterial community diversity of the Tucson samples was clearly distinguishable from that of New York and San Francisco, which were indistinguishable. Overall, our comprehensive molecular analysis of office building microbial diversity shows the potential of these methods for studying patterns and origins of indoor bacterial contamination. Humans move through a sea of microbial life that is seldom perceived except in the context of potential disease and decay Feazel et al. (2009)."	ENVO:surface	NA	Office Space Bacterial Abundance and Diversity in Three  Metropolitan Areas	NA	NA	Office Map study	FWD:AGAGTTTGATCCTGGCTCAG; REV:TGCTGCCTCCCGTAGGAGT	22666400	NA	NA	NA	Dan Knights/Justin Kuczinsky	CCME	GAZ:United States of America	NA	NA	0	NA	NA	NA	NA	n	NA	Scott Kelley	Kelley_office_contamination	pyrosequencing	NA	NA	NA	metagenomes	NA	ENVO:terrestrial biome	FLX	NA	NA	NA	NA	NA	Kelley_office_contamination	Kelley_Office_Map	408169	CCME	NA	NA	NA	1479	FLX protocol	NA	NA	n	NA	NA	ENVO:surface	16S rRNA	NA	NA	V2	CCME	NA	Chair	NA	NA	NA	32.22174	NA	NA	Swab_sample_from_woman_chair_in_Tucson_office_room_22
TUFC24.522438	CGTACTAGACTG	CATGCTGCCTCCCGTAGGAGT	NA	NA	Tucson	Chair	NA	NA	Chair	y	2011	Kelley_Office_Map	NA	FWIRNKE01	NA	mimarks-survey	NA	NA	NA	NA	0	NA	NA	758.65747	2011	NA	0	NA	NA	female	NA	NA	NA	drscottkelley@gmail.com	NA	24	NA	".1,g"	NA	NA	"Samples from office surfaces in New York, San Francisco and Tucson"	NA	NA	-110.9264	y	NA	CCME	NA	NA	NA	"People in developed countries spend approximately 90% of their lives indoors, yet we know little about the source and diversity of microbes in built environments. In this study, we combined culture-based cell counting and multiplexed pyrosequencing of environmental ribosomal RNA (rRNA) gene sequences to investigate office space bacterial diversity in three metropolitan areas. Five surfaces common to all offices were sampled using sterile double-tipped swabs, one tip for culturing and one for DNA extraction, in 30 different offices per city (90 offices, 450 total samples). 16S rRNA gene sequences were PCR amplified using bar-coded universal bacterial primers from 54 of the surfaces (18 per city) and pooled for pyrosequencing. A three-factorial Analysis of Variance (ANOVA) found significant differences in viable bacterial abundance between offices inhabited by men or women, among the various surface types, and among cities. Multiplex pyrosequencing identified more than 500 bacterial genera from 20 different bacterial divisions. The most abundant of these genera tended to be common inhabitants of human skin, nasal, oral or intestinal cavities. Other commonly occurring genera appeared to have environmental origins (e.g., soils). There were no significant differences in the bacterial diversity between offices inhabited by men or women or among surfaces, but the bacterial community diversity of the Tucson samples was clearly distinguishable from that of New York and San Francisco, which were indistinguishable. Overall, our comprehensive molecular analysis of office building microbial diversity shows the potential of these methods for studying patterns and origins of indoor bacterial contamination. Humans move through a sea of microbial life that is seldom perceived except in the context of potential disease and decay Feazel et al. (2009)."	ENVO:surface	NA	Office Space Bacterial Abundance and Diversity in Three  Metropolitan Areas	NA	NA	Office Map study	FWD:AGAGTTTGATCCTGGCTCAG; REV:TGCTGCCTCCCGTAGGAGT	22666400	NA	NA	NA	Dan Knights/Justin Kuczinsky	CCME	GAZ:United States of America	NA	NA	0	NA	NA	NA	NA	n	NA	Scott Kelley	Kelley_office_contamination	pyrosequencing	NA	NA	NA	metagenomes	NA	ENVO:terrestrial biome	FLX	NA	NA	NA	NA	NA	Kelley_office_contamination	Kelley_Office_Map	408169	CCME	NA	NA	NA	1479	FLX protocol	NA	NA	n	NA	NA	ENVO:surface	16S rRNA	NA	NA	V2	CCME	NA	Chair	NA	NA	NA	32.22174	NA	NA	Swab_sample_from_woman_chair_in_Tucson_office_room_24
TUFC4.522468	CGTCAACGATGT	CATGCTGCCTCCCGTAGGAGT	NA	NA	Tucson	Chair	NA	NA	Chair	y	2011	Kelley_Office_Map	NA	FWIRNKE01	NA	mimarks-survey	NA	NA	NA	NA	0	NA	NA	758.65747	2011	NA	0	NA	NA	female	NA	NA	NA	drscottkelley@gmail.com	NA	4	NA	".1,g"	NA	NA	"Samples from office surfaces in New York, San Francisco and Tucson"	NA	NA	-110.9264	y	NA	CCME	NA	NA	NA	"People in developed countries spend approximately 90% of their lives indoors, yet we know little about the source and diversity of microbes in built environments. In this study, we combined culture-based cell counting and multiplexed pyrosequencing of environmental ribosomal RNA (rRNA) gene sequences to investigate office space bacterial diversity in three metropolitan areas. Five surfaces common to all offices were sampled using sterile double-tipped swabs, one tip for culturing and one for DNA extraction, in 30 different offices per city (90 offices, 450 total samples). 16S rRNA gene sequences were PCR amplified using bar-coded universal bacterial primers from 54 of the surfaces (18 per city) and pooled for pyrosequencing. A three-factorial Analysis of Variance (ANOVA) found significant differences in viable bacterial abundance between offices inhabited by men or women, among the various surface types, and among cities. Multiplex pyrosequencing identified more than 500 bacterial genera from 20 different bacterial divisions. The most abundant of these genera tended to be common inhabitants of human skin, nasal, oral or intestinal cavities. Other commonly occurring genera appeared to have environmental origins (e.g., soils). There were no significant differences in the bacterial diversity between offices inhabited by men or women or among surfaces, but the bacterial community diversity of the Tucson samples was clearly distinguishable from that of New York and San Francisco, which were indistinguishable. Overall, our comprehensive molecular analysis of office building microbial diversity shows the potential of these methods for studying patterns and origins of indoor bacterial contamination. Humans move through a sea of microbial life that is seldom perceived except in the context of potential disease and decay Feazel et al. (2009)."	ENVO:surface	NA	Office Space Bacterial Abundance and Diversity in Three  Metropolitan Areas	NA	NA	Office Map study	FWD:AGAGTTTGATCCTGGCTCAG; REV:TGCTGCCTCCCGTAGGAGT	22666400	NA	NA	NA	Dan Knights/Justin Kuczinsky	CCME	GAZ:United States of America	NA	NA	0	NA	NA	NA	NA	n	NA	Scott Kelley	Kelley_office_contamination	pyrosequencing	NA	NA	NA	metagenomes	NA	ENVO:terrestrial biome	FLX	NA	NA	NA	NA	NA	Kelley_office_contamination	Kelley_Office_Map	408169	CCME	NA	NA	NA	1479	FLX protocol	NA	NA	n	NA	NA	ENVO:surface	16S rRNA	NA	NA	V2	CCME	NA	Chair	NA	NA	NA	32.22174	NA	NA	Swab_sample_from_woman_chair_in_Tucson_office_room_4
TUMC11.522440	CGCGATAGCAGT	CATGCTGCCTCCCGTAGGAGT	NA	NA	Tucson	Chair	NA	NA	Chair	y	2011	Kelley_Office_Map	NA	FWIRNKE01	NA	mimarks-survey	NA	NA	NA	NA	0	NA	NA	758.65747	2011	NA	0	NA	NA	male	NA	NA	NA	drscottkelley@gmail.com	NA	11	NA	".1,g"	NA	NA	"Samples from office surfaces in New York, San Francisco and Tucson"	NA	NA	-110.9264	y	NA	CCME	NA	NA	NA	"People in developed countries spend approximately 90% of their lives indoors, yet we know little about the source and diversity of microbes in built environments. In this study, we combined culture-based cell counting and multiplexed pyrosequencing of environmental ribosomal RNA (rRNA) gene sequences to investigate office space bacterial diversity in three metropolitan areas. Five surfaces common to all offices were sampled using sterile double-tipped swabs, one tip for culturing and one for DNA extraction, in 30 different offices per city (90 offices, 450 total samples). 16S rRNA gene sequences were PCR amplified using bar-coded universal bacterial primers from 54 of the surfaces (18 per city) and pooled for pyrosequencing. A three-factorial Analysis of Variance (ANOVA) found significant differences in viable bacterial abundance between offices inhabited by men or women, among the various surface types, and among cities. Multiplex pyrosequencing identified more than 500 bacterial genera from 20 different bacterial divisions. The most abundant of these genera tended to be common inhabitants of human skin, nasal, oral or intestinal cavities. Other commonly occurring genera appeared to have environmental origins (e.g., soils). There were no significant differences in the bacterial diversity between offices inhabited by men or women or among surfaces, but the bacterial community diversity of the Tucson samples was clearly distinguishable from that of New York and San Francisco, which were indistinguishable. Overall, our comprehensive molecular analysis of office building microbial diversity shows the potential of these methods for studying patterns and origins of indoor bacterial contamination. Humans move through a sea of microbial life that is seldom perceived except in the context of potential disease and decay Feazel et al. (2009)."	ENVO:surface	NA	Office Space Bacterial Abundance and Diversity in Three  Metropolitan Areas	NA	NA	Office Map study	FWD:AGAGTTTGATCCTGGCTCAG; REV:TGCTGCCTCCCGTAGGAGT	22666400	NA	NA	NA	Dan Knights/Justin Kuczinsky	CCME	GAZ:United States of America	NA	NA	0	NA	NA	NA	NA	n	NA	Scott Kelley	Kelley_office_contamination	pyrosequencing	NA	NA	NA	metagenomes	NA	ENVO:terrestrial biome	FLX	NA	NA	NA	NA	NA	Kelley_office_contamination	Kelley_Office_Map	408169	CCME	NA	NA	NA	1479	FLX protocol	NA	NA	n	NA	NA	ENVO:surface	16S rRNA	NA	NA	V2	CCME	NA	Chair	NA	NA	NA	32.22174	NA	NA	Swab_sample_from_man_chair_in_Tucson_office_room_11
TUMC15.522483	CGCGTAACTGTA	CATGCTGCCTCCCGTAGGAGT	NA	NA	Tucson	Chair	NA	NA	Chair	y	2011	Kelley_Office_Map	NA	FWIRNKE01	NA	mimarks-survey	NA	NA	NA	NA	0	NA	NA	758.65747	2011	NA	0	NA	NA	male	NA	NA	NA	drscottkelley@gmail.com	NA	15	NA	".1,g"	NA	NA	"Samples from office surfaces in New York, San Francisco and Tucson"	NA	NA	-110.9264	y	NA	CCME	NA	NA	NA	"People in developed countries spend approximately 90% of their lives indoors, yet we know little about the source and diversity of microbes in built environments. In this study, we combined culture-based cell counting and multiplexed pyrosequencing of environmental ribosomal RNA (rRNA) gene sequences to investigate office space bacterial diversity in three metropolitan areas. Five surfaces common to all offices were sampled using sterile double-tipped swabs, one tip for culturing and one for DNA extraction, in 30 different offices per city (90 offices, 450 total samples). 16S rRNA gene sequences were PCR amplified using bar-coded universal bacterial primers from 54 of the surfaces (18 per city) and pooled for pyrosequencing. A three-factorial Analysis of Variance (ANOVA) found significant differences in viable bacterial abundance between offices inhabited by men or women, among the various surface types, and among cities. Multiplex pyrosequencing identified more than 500 bacterial genera from 20 different bacterial divisions. The most abundant of these genera tended to be common inhabitants of human skin, nasal, oral or intestinal cavities. Other commonly occurring genera appeared to have environmental origins (e.g., soils). There were no significant differences in the bacterial diversity between offices inhabited by men or women or among surfaces, but the bacterial community diversity of the Tucson samples was clearly distinguishable from that of New York and San Francisco, which were indistinguishable. Overall, our comprehensive molecular analysis of office building microbial diversity shows the potential of these methods for studying patterns and origins of indoor bacterial contamination. Humans move through a sea of microbial life that is seldom perceived except in the context of potential disease and decay Feazel et al. (2009)."	ENVO:surface	NA	Office Space Bacterial Abundance and Diversity in Three  Metropolitan Areas	NA	NA	Office Map study	FWD:AGAGTTTGATCCTGGCTCAG; REV:TGCTGCCTCCCGTAGGAGT	22666400	NA	NA	NA	Dan Knights/Justin Kuczinsky	CCME	GAZ:United States of America	NA	NA	0	NA	NA	NA	NA	n	NA	Scott Kelley	Kelley_office_contamination	pyrosequencing	NA	NA	NA	metagenomes	NA	ENVO:terrestrial biome	FLX	NA	NA	NA	NA	NA	Kelley_office_contamination	Kelley_Office_Map	408169	CCME	NA	NA	NA	1479	FLX protocol	NA	NA	n	NA	NA	ENVO:surface	16S rRNA	NA	NA	V2	CCME	NA	Chair	NA	NA	NA	32.22174	NA	NA	Swab_sample_from_man_chair_in_Tucson_office_room_15
TUMC18.522481	CGCTAGAACGCA	CATGCTGCCTCCCGTAGGAGT	NA	NA	Tucson	Chair	NA	NA	Chair	y	2011	Kelley_Office_Map	NA	FWIRNKE01	NA	mimarks-survey	NA	NA	NA	NA	0	NA	NA	758.65747	2011	NA	0	NA	NA	male	NA	NA	NA	drscottkelley@gmail.com	NA	18	NA	".1,g"	NA	NA	"Samples from office surfaces in New York, San Francisco and Tucson"	NA	NA	-110.9264	y	NA	CCME	NA	NA	NA	"People in developed countries spend approximately 90% of their lives indoors, yet we know little about the source and diversity of microbes in built environments. In this study, we combined culture-based cell counting and multiplexed pyrosequencing of environmental ribosomal RNA (rRNA) gene sequences to investigate office space bacterial diversity in three metropolitan areas. Five surfaces common to all offices were sampled using sterile double-tipped swabs, one tip for culturing and one for DNA extraction, in 30 different offices per city (90 offices, 450 total samples). 16S rRNA gene sequences were PCR amplified using bar-coded universal bacterial primers from 54 of the surfaces (18 per city) and pooled for pyrosequencing. A three-factorial Analysis of Variance (ANOVA) found significant differences in viable bacterial abundance between offices inhabited by men or women, among the various surface types, and among cities. Multiplex pyrosequencing identified more than 500 bacterial genera from 20 different bacterial divisions. The most abundant of these genera tended to be common inhabitants of human skin, nasal, oral or intestinal cavities. Other commonly occurring genera appeared to have environmental origins (e.g., soils). There were no significant differences in the bacterial diversity between offices inhabited by men or women or among surfaces, but the bacterial community diversity of the Tucson samples was clearly distinguishable from that of New York and San Francisco, which were indistinguishable. Overall, our comprehensive molecular analysis of office building microbial diversity shows the potential of these methods for studying patterns and origins of indoor bacterial contamination. Humans move through a sea of microbial life that is seldom perceived except in the context of potential disease and decay Feazel et al. (2009)."	ENVO:surface	NA	Office Space Bacterial Abundance and Diversity in Three  Metropolitan Areas	NA	NA	Office Map study	FWD:AGAGTTTGATCCTGGCTCAG; REV:TGCTGCCTCCCGTAGGAGT	22666400	NA	NA	NA	Dan Knights/Justin Kuczinsky	CCME	GAZ:United States of America	NA	NA	0	NA	NA	NA	NA	n	NA	Scott Kelley	Kelley_office_contamination	pyrosequencing	NA	NA	NA	metagenomes	NA	ENVO:terrestrial biome	FLX	NA	NA	NA	NA	NA	Kelley_office_contamination	Kelley_Office_Map	408169	CCME	NA	NA	NA	1479	FLX protocol	NA	NA	n	NA	NA	ENVO:surface	16S rRNA	NA	NA	V2	CCME	NA	Chair	NA	NA	NA	32.22174	NA	NA	Swab_sample_from_man_chair_in_Tucson_office_room_18
TUMC23.522480	CGCTTATCGAGA	CATGCTGCCTCCCGTAGGAGT	NA	NA	Tucson	Chair	NA	NA	Chair	y	2011	Kelley_Office_Map	NA	FWIRNKE01	NA	mimarks-survey	NA	NA	NA	NA	0	NA	NA	758.65747	2011	NA	0	NA	NA	male	NA	NA	NA	drscottkelley@gmail.com	NA	23	NA	".1,g"	NA	NA	"Samples from office surfaces in New York, San Francisco and Tucson"	NA	NA	-110.9264	y	NA	CCME	NA	NA	NA	"People in developed countries spend approximately 90% of their lives indoors, yet we know little about the source and diversity of microbes in built environments. In this study, we combined culture-based cell counting and multiplexed pyrosequencing of environmental ribosomal RNA (rRNA) gene sequences to investigate office space bacterial diversity in three metropolitan areas. Five surfaces common to all offices were sampled using sterile double-tipped swabs, one tip for culturing and one for DNA extraction, in 30 different offices per city (90 offices, 450 total samples). 16S rRNA gene sequences were PCR amplified using bar-coded universal bacterial primers from 54 of the surfaces (18 per city) and pooled for pyrosequencing. A three-factorial Analysis of Variance (ANOVA) found significant differences in viable bacterial abundance between offices inhabited by men or women, among the various surface types, and among cities. Multiplex pyrosequencing identified more than 500 bacterial genera from 20 different bacterial divisions. The most abundant of these genera tended to be common inhabitants of human skin, nasal, oral or intestinal cavities. Other commonly occurring genera appeared to have environmental origins (e.g., soils). There were no significant differences in the bacterial diversity between offices inhabited by men or women or among surfaces, but the bacterial community diversity of the Tucson samples was clearly distinguishable from that of New York and San Francisco, which were indistinguishable. Overall, our comprehensive molecular analysis of office building microbial diversity shows the potential of these methods for studying patterns and origins of indoor bacterial contamination. Humans move through a sea of microbial life that is seldom perceived except in the context of potential disease and decay Feazel et al. (2009)."	ENVO:surface	NA	Office Space Bacterial Abundance and Diversity in Three  Metropolitan Areas	NA	NA	Office Map study	FWD:AGAGTTTGATCCTGGCTCAG; REV:TGCTGCCTCCCGTAGGAGT	22666400	NA	NA	NA	Dan Knights/Justin Kuczinsky	CCME	GAZ:United States of America	NA	NA	0	NA	NA	NA	NA	n	NA	Scott Kelley	Kelley_office_contamination	pyrosequencing	NA	NA	NA	metagenomes	NA	ENVO:terrestrial biome	FLX	NA	NA	NA	NA	NA	Kelley_office_contamination	Kelley_Office_Map	408169	CCME	NA	NA	NA	1479	FLX protocol	NA	NA	n	NA	NA	ENVO:surface	16S rRNA	NA	NA	V2	CCME	NA	Chair	NA	NA	NA	32.22174	NA	NA	Swab_sample_from_man_chair_in_Tucson_office_room_23
TUMC25.522469	CGGAGTGTCTAT	CATGCTGCCTCCCGTAGGAGT	NA	NA	Tucson	Chair	NA	NA	Chair	y	2011	Kelley_Office_Map	NA	FWIRNKE01	NA	mimarks-survey	NA	NA	NA	NA	0	NA	NA	758.65747	2011	NA	0	NA	NA	male	NA	NA	NA	drscottkelley@gmail.com	NA	25	NA	".1,g"	NA	NA	"Samples from office surfaces in New York, San Francisco and Tucson"	NA	NA	-110.9264	y	NA	CCME	NA	NA	NA	"People in developed countries spend approximately 90% of their lives indoors, yet we know little about the source and diversity of microbes in built environments. In this study, we combined culture-based cell counting and multiplexed pyrosequencing of environmental ribosomal RNA (rRNA) gene sequences to investigate office space bacterial diversity in three metropolitan areas. Five surfaces common to all offices were sampled using sterile double-tipped swabs, one tip for culturing and one for DNA extraction, in 30 different offices per city (90 offices, 450 total samples). 16S rRNA gene sequences were PCR amplified using bar-coded universal bacterial primers from 54 of the surfaces (18 per city) and pooled for pyrosequencing. A three-factorial Analysis of Variance (ANOVA) found significant differences in viable bacterial abundance between offices inhabited by men or women, among the various surface types, and among cities. Multiplex pyrosequencing identified more than 500 bacterial genera from 20 different bacterial divisions. The most abundant of these genera tended to be common inhabitants of human skin, nasal, oral or intestinal cavities. Other commonly occurring genera appeared to have environmental origins (e.g., soils). There were no significant differences in the bacterial diversity between offices inhabited by men or women or among surfaces, but the bacterial community diversity of the Tucson samples was clearly distinguishable from that of New York and San Francisco, which were indistinguishable. Overall, our comprehensive molecular analysis of office building microbial diversity shows the potential of these methods for studying patterns and origins of indoor bacterial contamination. Humans move through a sea of microbial life that is seldom perceived except in the context of potential disease and decay Feazel et al. (2009)."	ENVO:surface	NA	Office Space Bacterial Abundance and Diversity in Three  Metropolitan Areas	NA	NA	Office Map study	FWD:AGAGTTTGATCCTGGCTCAG; REV:TGCTGCCTCCCGTAGGAGT	22666400	NA	NA	NA	Dan Knights/Justin Kuczinsky	CCME	GAZ:United States of America	NA	NA	0	NA	NA	NA	NA	n	NA	Scott Kelley	Kelley_office_contamination	pyrosequencing	NA	NA	NA	metagenomes	NA	ENVO:terrestrial biome	FLX	NA	NA	NA	NA	NA	Kelley_office_contamination	Kelley_Office_Map	408169	CCME	NA	NA	NA	1479	FLX protocol	NA	NA	n	NA	NA	ENVO:surface	16S rRNA	NA	NA	V2	CCME	NA	Chair	NA	NA	NA	32.22174	NA	NA	Swab_sample_from_man_chair_in_Tucson_office_room_25
NYFP16.522442	TCATGGTACACT	CATGCTGCCTCCCGTAGGAGT	NA	NA	NewYork	Phone	NA	NA	Phone	y	2011	Kelley_Office_Map	NA	FWIRNKE01	NA	mimarks-survey	NA	NA	NA	NA	0	NA	NA	9.777	2011	NA	0	NA	NA	female	NA	NA	NA	drscottkelley@gmail.com	NA	16	NA	".1,g"	NA	NA	"Samples from office surfaces in New York, San Francisco and Tucson"	NA	NA	-74.00597	y	NA	CCME	NA	NA	NA	"People in developed countries spend approximately 90% of their lives indoors, yet we know little about the source and diversity of microbes in built environments. In this study, we combined culture-based cell counting and multiplexed pyrosequencing of environmental ribosomal RNA (rRNA) gene sequences to investigate office space bacterial diversity in three metropolitan areas. Five surfaces common to all offices were sampled using sterile double-tipped swabs, one tip for culturing and one for DNA extraction, in 30 different offices per city (90 offices, 450 total samples). 16S rRNA gene sequences were PCR amplified using bar-coded universal bacterial primers from 54 of the surfaces (18 per city) and pooled for pyrosequencing. A three-factorial Analysis of Variance (ANOVA) found significant differences in viable bacterial abundance between offices inhabited by men or women, among the various surface types, and among cities. Multiplex pyrosequencing identified more than 500 bacterial genera from 20 different bacterial divisions. The most abundant of these genera tended to be common inhabitants of human skin, nasal, oral or intestinal cavities. Other commonly occurring genera appeared to have environmental origins (e.g., soils). There were no significant differences in the bacterial diversity between offices inhabited by men or women or among surfaces, but the bacterial community diversity of the Tucson samples was clearly distinguishable from that of New York and San Francisco, which were indistinguishable. Overall, our comprehensive molecular analysis of office building microbial diversity shows the potential of these methods for studying patterns and origins of indoor bacterial contamination. Humans move through a sea of microbial life that is seldom perceived except in the context of potential disease and decay Feazel et al. (2009)."	ENVO:surface	NA	Office Space Bacterial Abundance and Diversity in Three  Metropolitan Areas	NA	NA	Office Map study	FWD:AGAGTTTGATCCTGGCTCAG; REV:TGCTGCCTCCCGTAGGAGT	22666400	NA	NA	NA	Dan Knights/Justin Kuczinsky	CCME	GAZ:United States of America	NA	NA	0	NA	NA	NA	NA	n	NA	Scott Kelley	Kelley_office_contamination	pyrosequencing	NA	NA	NA	metagenomes	NA	ENVO:terrestrial biome	FLX	NA	NA	NA	NA	NA	Kelley_office_contamination	Kelley_Office_Map	408169	CCME	NA	NA	NA	1479	FLX protocol	NA	NA	n	NA	NA	ENVO:surface	16S rRNA	NA	NA	V2	CCME	NA	Phone	NA	NA	NA	40.71435	NA	NA	Swab_sample_from_woman_phone_in_New_York_office_room_16
NYFP19.522441	TCATCGCGATAT	CATGCTGCCTCCCGTAGGAGT	NA	NA	NewYork	Phone	NA	NA	Phone	y	2011	Kelley_Office_Map	NA	FWIRNKE01	NA	mimarks-survey	NA	NA	NA	NA	0	NA	NA	9.777	2011	NA	0	NA	NA	female	NA	NA	NA	drscottkelley@gmail.com	NA	19	NA	".1,g"	NA	NA	"Samples from office surfaces in New York, San Francisco and Tucson"	NA	NA	-74.00597	y	NA	CCME	NA	NA	NA	"People in developed countries spend approximately 90% of their lives indoors, yet we know little about the source and diversity of microbes in built environments. In this study, we combined culture-based cell counting and multiplexed pyrosequencing of environmental ribosomal RNA (rRNA) gene sequences to investigate office space bacterial diversity in three metropolitan areas. Five surfaces common to all offices were sampled using sterile double-tipped swabs, one tip for culturing and one for DNA extraction, in 30 different offices per city (90 offices, 450 total samples). 16S rRNA gene sequences were PCR amplified using bar-coded universal bacterial primers from 54 of the surfaces (18 per city) and pooled for pyrosequencing. A three-factorial Analysis of Variance (ANOVA) found significant differences in viable bacterial abundance between offices inhabited by men or women, among the various surface types, and among cities. Multiplex pyrosequencing identified more than 500 bacterial genera from 20 different bacterial divisions. The most abundant of these genera tended to be common inhabitants of human skin, nasal, oral or intestinal cavities. Other commonly occurring genera appeared to have environmental origins (e.g., soils). There were no significant differences in the bacterial diversity between offices inhabited by men or women or among surfaces, but the bacterial community diversity of the Tucson samples was clearly distinguishable from that of New York and San Francisco, which were indistinguishable. Overall, our comprehensive molecular analysis of office building microbial diversity shows the potential of these methods for studying patterns and origins of indoor bacterial contamination. Humans move through a sea of microbial life that is seldom perceived except in the context of potential disease and decay Feazel et al. (2009)."	ENVO:surface	NA	Office Space Bacterial Abundance and Diversity in Three  Metropolitan Areas	NA	NA	Office Map study	FWD:AGAGTTTGATCCTGGCTCAG; REV:TGCTGCCTCCCGTAGGAGT	22666400	NA	NA	NA	Dan Knights/Justin Kuczinsky	CCME	GAZ:United States of America	NA	NA	0	NA	NA	NA	NA	n	NA	Scott Kelley	Kelley_office_contamination	pyrosequencing	NA	NA	NA	metagenomes	NA	ENVO:terrestrial biome	FLX	NA	NA	NA	NA	NA	Kelley_office_contamination	Kelley_Office_Map	408169	CCME	NA	NA	NA	1479	FLX protocol	NA	NA	n	NA	NA	ENVO:surface	16S rRNA	NA	NA	V2	CCME	NA	Phone	NA	NA	NA	40.71435	NA	NA	Swab_sample_from_woman_phone_in_New_York_office_room_19
NYFP22.522470	TCATCTGACTGA	CATGCTGCCTCCCGTAGGAGT	NA	NA	NewYork	Phone	NA	NA	Phone	y	2011	Kelley_Office_Map	NA	FWIRNKE01	NA	mimarks-survey	NA	NA	NA	NA	0	NA	NA	9.777	2011	NA	0	NA	NA	female	NA	NA	NA	drscottkelley@gmail.com	NA	22	NA	".1,g"	NA	NA	"Samples from office surfaces in New York, San Francisco and Tucson"	NA	NA	-74.00597	y	NA	CCME	NA	NA	NA	"People in developed countries spend approximately 90% of their lives indoors, yet we know little about the source and diversity of microbes in built environments. In this study, we combined culture-based cell counting and multiplexed pyrosequencing of environmental ribosomal RNA (rRNA) gene sequences to investigate office space bacterial diversity in three metropolitan areas. Five surfaces common to all offices were sampled using sterile double-tipped swabs, one tip for culturing and one for DNA extraction, in 30 different offices per city (90 offices, 450 total samples). 16S rRNA gene sequences were PCR amplified using bar-coded universal bacterial primers from 54 of the surfaces (18 per city) and pooled for pyrosequencing. A three-factorial Analysis of Variance (ANOVA) found significant differences in viable bacterial abundance between offices inhabited by men or women, among the various surface types, and among cities. Multiplex pyrosequencing identified more than 500 bacterial genera from 20 different bacterial divisions. The most abundant of these genera tended to be common inhabitants of human skin, nasal, oral or intestinal cavities. Other commonly occurring genera appeared to have environmental origins (e.g., soils). There were no significant differences in the bacterial diversity between offices inhabited by men or women or among surfaces, but the bacterial community diversity of the Tucson samples was clearly distinguishable from that of New York and San Francisco, which were indistinguishable. Overall, our comprehensive molecular analysis of office building microbial diversity shows the potential of these methods for studying patterns and origins of indoor bacterial contamination. Humans move through a sea of microbial life that is seldom perceived except in the context of potential disease and decay Feazel et al. (2009)."	ENVO:surface	NA	Office Space Bacterial Abundance and Diversity in Three  Metropolitan Areas	NA	NA	Office Map study	FWD:AGAGTTTGATCCTGGCTCAG; REV:TGCTGCCTCCCGTAGGAGT	22666400	NA	NA	NA	Dan Knights/Justin Kuczinsky	CCME	GAZ:United States of America	NA	NA	0	NA	NA	NA	NA	n	NA	Scott Kelley	Kelley_office_contamination	pyrosequencing	NA	NA	NA	metagenomes	NA	ENVO:terrestrial biome	FLX	NA	NA	NA	NA	NA	Kelley_office_contamination	Kelley_Office_Map	408169	CCME	NA	NA	NA	1479	FLX protocol	NA	NA	n	NA	NA	ENVO:surface	16S rRNA	NA	NA	V2	CCME	NA	Phone	NA	NA	NA	40.71435	NA	NA	Swab_sample_from_woman_phone_in_New_York_office_room_22
NYFP4.522464	TCAGTCGACGAG	CATGCTGCCTCCCGTAGGAGT	NA	NA	NewYork	Phone	NA	NA	Phone	y	2011	Kelley_Office_Map	NA	FWIRNKE01	NA	mimarks-survey	NA	NA	NA	NA	0	NA	NA	9.777	2011	NA	0	NA	NA	female	NA	NA	NA	drscottkelley@gmail.com	NA	4	NA	".1,g"	NA	NA	"Samples from office surfaces in New York, San Francisco and Tucson"	NA	NA	-74.00597	y	NA	CCME	NA	NA	NA	"People in developed countries spend approximately 90% of their lives indoors, yet we know little about the source and diversity of microbes in built environments. In this study, we combined culture-based cell counting and multiplexed pyrosequencing of environmental ribosomal RNA (rRNA) gene sequences to investigate office space bacterial diversity in three metropolitan areas. Five surfaces common to all offices were sampled using sterile double-tipped swabs, one tip for culturing and one for DNA extraction, in 30 different offices per city (90 offices, 450 total samples). 16S rRNA gene sequences were PCR amplified using bar-coded universal bacterial primers from 54 of the surfaces (18 per city) and pooled for pyrosequencing. A three-factorial Analysis of Variance (ANOVA) found significant differences in viable bacterial abundance between offices inhabited by men or women, among the various surface types, and among cities. Multiplex pyrosequencing identified more than 500 bacterial genera from 20 different bacterial divisions. The most abundant of these genera tended to be common inhabitants of human skin, nasal, oral or intestinal cavities. Other commonly occurring genera appeared to have environmental origins (e.g., soils). There were no significant differences in the bacterial diversity between offices inhabited by men or women or among surfaces, but the bacterial community diversity of the Tucson samples was clearly distinguishable from that of New York and San Francisco, which were indistinguishable. Overall, our comprehensive molecular analysis of office building microbial diversity shows the potential of these methods for studying patterns and origins of indoor bacterial contamination. Humans move through a sea of microbial life that is seldom perceived except in the context of potential disease and decay Feazel et al. (2009)."	ENVO:surface	NA	Office Space Bacterial Abundance and Diversity in Three  Metropolitan Areas	NA	NA	Office Map study	FWD:AGAGTTTGATCCTGGCTCAG; REV:TGCTGCCTCCCGTAGGAGT	22666400	NA	NA	NA	Dan Knights/Justin Kuczinsky	CCME	GAZ:United States of America	NA	NA	0	NA	NA	NA	NA	n	NA	Scott Kelley	Kelley_office_contamination	pyrosequencing	NA	NA	NA	metagenomes	NA	ENVO:terrestrial biome	FLX	NA	NA	NA	NA	NA	Kelley_office_contamination	Kelley_Office_Map	408169	CCME	NA	NA	NA	1479	FLX protocol	NA	NA	n	NA	NA	ENVO:surface	16S rRNA	NA	NA	V2	CCME	NA	Phone	NA	NA	NA	40.71435	NA	NA	Swab_sample_from_woman_phone_in_New_York_office_room_4
NYFP7.522477	TCAGTGACGTAC	CATGCTGCCTCCCGTAGGAGT	NA	NA	NewYork	Phone	NA	NA	Phone	y	2011	Kelley_Office_Map	NA	FWIRNKE01	NA	mimarks-survey	NA	NA	NA	NA	0	NA	NA	9.777	2011	NA	0	NA	NA	female	NA	NA	NA	drscottkelley@gmail.com	NA	7	NA	".1,g"	NA	NA	"Samples from office surfaces in New York, San Francisco and Tucson"	NA	NA	-74.00597	y	NA	CCME	NA	NA	NA	"People in developed countries spend approximately 90% of their lives indoors, yet we know little about the source and diversity of microbes in built environments. In this study, we combined culture-based cell counting and multiplexed pyrosequencing of environmental ribosomal RNA (rRNA) gene sequences to investigate office space bacterial diversity in three metropolitan areas. Five surfaces common to all offices were sampled using sterile double-tipped swabs, one tip for culturing and one for DNA extraction, in 30 different offices per city (90 offices, 450 total samples). 16S rRNA gene sequences were PCR amplified using bar-coded universal bacterial primers from 54 of the surfaces (18 per city) and pooled for pyrosequencing. A three-factorial Analysis of Variance (ANOVA) found significant differences in viable bacterial abundance between offices inhabited by men or women, among the various surface types, and among cities. Multiplex pyrosequencing identified more than 500 bacterial genera from 20 different bacterial divisions. The most abundant of these genera tended to be common inhabitants of human skin, nasal, oral or intestinal cavities. Other commonly occurring genera appeared to have environmental origins (e.g., soils). There were no significant differences in the bacterial diversity between offices inhabited by men or women or among surfaces, but the bacterial community diversity of the Tucson samples was clearly distinguishable from that of New York and San Francisco, which were indistinguishable. Overall, our comprehensive molecular analysis of office building microbial diversity shows the potential of these methods for studying patterns and origins of indoor bacterial contamination. Humans move through a sea of microbial life that is seldom perceived except in the context of potential disease and decay Feazel et al. (2009)."	ENVO:surface	NA	Office Space Bacterial Abundance and Diversity in Three  Metropolitan Areas	NA	NA	Office Map study	FWD:AGAGTTTGATCCTGGCTCAG; REV:TGCTGCCTCCCGTAGGAGT	22666400	NA	NA	NA	Dan Knights/Justin Kuczinsky	CCME	GAZ:United States of America	NA	NA	0	NA	NA	NA	NA	n	NA	Scott Kelley	Kelley_office_contamination	pyrosequencing	NA	NA	NA	metagenomes	NA	ENVO:terrestrial biome	FLX	NA	NA	NA	NA	NA	Kelley_office_contamination	Kelley_Office_Map	408169	CCME	NA	NA	NA	1479	FLX protocol	NA	NA	n	NA	NA	ENVO:surface	16S rRNA	NA	NA	V2	CCME	NA	Phone	NA	NA	NA	40.71435	NA	NA	Swab_sample_from_woman_phone_in_New_York_office_room_7
NYMP18.522453	TCACTATGGTCA	CATGCTGCCTCCCGTAGGAGT	NA	NA	NewYork	Phone	NA	NA	Phone	y	2011	Kelley_Office_Map	NA	FWIRNKE01	NA	mimarks-survey	NA	NA	NA	NA	0	NA	NA	9.777	2011	NA	0	NA	NA	male	NA	NA	NA	drscottkelley@gmail.com	NA	18	NA	".1,g"	NA	NA	"Samples from office surfaces in New York, San Francisco and Tucson"	NA	NA	-74.00597	y	NA	CCME	NA	NA	NA	"People in developed countries spend approximately 90% of their lives indoors, yet we know little about the source and diversity of microbes in built environments. In this study, we combined culture-based cell counting and multiplexed pyrosequencing of environmental ribosomal RNA (rRNA) gene sequences to investigate office space bacterial diversity in three metropolitan areas. Five surfaces common to all offices were sampled using sterile double-tipped swabs, one tip for culturing and one for DNA extraction, in 30 different offices per city (90 offices, 450 total samples). 16S rRNA gene sequences were PCR amplified using bar-coded universal bacterial primers from 54 of the surfaces (18 per city) and pooled for pyrosequencing. A three-factorial Analysis of Variance (ANOVA) found significant differences in viable bacterial abundance between offices inhabited by men or women, among the various surface types, and among cities. Multiplex pyrosequencing identified more than 500 bacterial genera from 20 different bacterial divisions. The most abundant of these genera tended to be common inhabitants of human skin, nasal, oral or intestinal cavities. Other commonly occurring genera appeared to have environmental origins (e.g., soils). There were no significant differences in the bacterial diversity between offices inhabited by men or women or among surfaces, but the bacterial community diversity of the Tucson samples was clearly distinguishable from that of New York and San Francisco, which were indistinguishable. Overall, our comprehensive molecular analysis of office building microbial diversity shows the potential of these methods for studying patterns and origins of indoor bacterial contamination. Humans move through a sea of microbial life that is seldom perceived except in the context of potential disease and decay Feazel et al. (2009)."	ENVO:surface	NA	Office Space Bacterial Abundance and Diversity in Three  Metropolitan Areas	NA	NA	Office Map study	FWD:AGAGTTTGATCCTGGCTCAG; REV:TGCTGCCTCCCGTAGGAGT	22666400	NA	NA	NA	Dan Knights/Justin Kuczinsky	CCME	GAZ:United States of America	NA	NA	0	NA	NA	NA	NA	n	NA	Scott Kelley	Kelley_office_contamination	pyrosequencing	NA	NA	NA	metagenomes	NA	ENVO:terrestrial biome	FLX	NA	NA	NA	NA	NA	Kelley_office_contamination	Kelley_Office_Map	408169	CCME	NA	NA	NA	1479	FLX protocol	NA	NA	n	NA	NA	ENVO:surface	16S rRNA	NA	NA	V2	CCME	NA	Phone	NA	NA	NA	40.71435	NA	NA	Swab_sample_from_man_phone_in_New_York_office_room_18
NYMP28.522461	TCACTTCTCGCT	CATGCTGCCTCCCGTAGGAGT	NA	NA	NewYork	Phone	NA	NA	Phone	y	2011	Kelley_Office_Map	NA	FWIRNKE01	NA	mimarks-survey	NA	NA	NA	NA	0	NA	NA	9.777	2011	NA	0	NA	NA	male	NA	NA	NA	drscottkelley@gmail.com	NA	28	NA	".1,g"	NA	NA	"Samples from office surfaces in New York, San Francisco and Tucson"	NA	NA	-74.00597	y	NA	CCME	NA	NA	NA	"People in developed countries spend approximately 90% of their lives indoors, yet we know little about the source and diversity of microbes in built environments. In this study, we combined culture-based cell counting and multiplexed pyrosequencing of environmental ribosomal RNA (rRNA) gene sequences to investigate office space bacterial diversity in three metropolitan areas. Five surfaces common to all offices were sampled using sterile double-tipped swabs, one tip for culturing and one for DNA extraction, in 30 different offices per city (90 offices, 450 total samples). 16S rRNA gene sequences were PCR amplified using bar-coded universal bacterial primers from 54 of the surfaces (18 per city) and pooled for pyrosequencing. A three-factorial Analysis of Variance (ANOVA) found significant differences in viable bacterial abundance between offices inhabited by men or women, among the various surface types, and among cities. Multiplex pyrosequencing identified more than 500 bacterial genera from 20 different bacterial divisions. The most abundant of these genera tended to be common inhabitants of human skin, nasal, oral or intestinal cavities. Other commonly occurring genera appeared to have environmental origins (e.g., soils). There were no significant differences in the bacterial diversity between offices inhabited by men or women or among surfaces, but the bacterial community diversity of the Tucson samples was clearly distinguishable from that of New York and San Francisco, which were indistinguishable. Overall, our comprehensive molecular analysis of office building microbial diversity shows the potential of these methods for studying patterns and origins of indoor bacterial contamination. Humans move through a sea of microbial life that is seldom perceived except in the context of potential disease and decay Feazel et al. (2009)."	ENVO:surface	NA	Office Space Bacterial Abundance and Diversity in Three  Metropolitan Areas	NA	NA	Office Map study	FWD:AGAGTTTGATCCTGGCTCAG; REV:TGCTGCCTCCCGTAGGAGT	22666400	NA	NA	NA	Dan Knights/Justin Kuczinsky	CCME	GAZ:United States of America	NA	NA	0	NA	NA	NA	NA	n	NA	Scott Kelley	Kelley_office_contamination	pyrosequencing	NA	NA	NA	metagenomes	NA	ENVO:terrestrial biome	FLX	NA	NA	NA	NA	NA	Kelley_office_contamination	Kelley_Office_Map	408169	CCME	NA	NA	NA	1479	FLX protocol	NA	NA	n	NA	NA	ENVO:surface	16S rRNA	NA	NA	V2	CCME	NA	Phone	NA	NA	NA	40.71435	NA	NA	Swab_sample_from_man_phone_in_New_York_office_room_28
NYMP31.522449	TCAGATCCGATG	CATGCTGCCTCCCGTAGGAGT	NA	NA	NewYork	Phone	NA	NA	Phone	y	2011	Kelley_Office_Map	NA	FWIRNKE01	NA	mimarks-survey	NA	NA	NA	NA	0	NA	NA	9.777	2011	NA	0	NA	NA	male	NA	NA	NA	drscottkelley@gmail.com	NA	31	NA	".1,g"	NA	NA	"Samples from office surfaces in New York, San Francisco and Tucson"	NA	NA	-74.00597	y	NA	CCME	NA	NA	NA	"People in developed countries spend approximately 90% of their lives indoors, yet we know little about the source and diversity of microbes in built environments. In this study, we combined culture-based cell counting and multiplexed pyrosequencing of environmental ribosomal RNA (rRNA) gene sequences to investigate office space bacterial diversity in three metropolitan areas. Five surfaces common to all offices were sampled using sterile double-tipped swabs, one tip for culturing and one for DNA extraction, in 30 different offices per city (90 offices, 450 total samples). 16S rRNA gene sequences were PCR amplified using bar-coded universal bacterial primers from 54 of the surfaces (18 per city) and pooled for pyrosequencing. A three-factorial Analysis of Variance (ANOVA) found significant differences in viable bacterial abundance between offices inhabited by men or women, among the various surface types, and among cities. Multiplex pyrosequencing identified more than 500 bacterial genera from 20 different bacterial divisions. The most abundant of these genera tended to be common inhabitants of human skin, nasal, oral or intestinal cavities. Other commonly occurring genera appeared to have environmental origins (e.g., soils). There were no significant differences in the bacterial diversity between offices inhabited by men or women or among surfaces, but the bacterial community diversity of the Tucson samples was clearly distinguishable from that of New York and San Francisco, which were indistinguishable. Overall, our comprehensive molecular analysis of office building microbial diversity shows the potential of these methods for studying patterns and origins of indoor bacterial contamination. Humans move through a sea of microbial life that is seldom perceived except in the context of potential disease and decay Feazel et al. (2009)."	ENVO:surface	NA	Office Space Bacterial Abundance and Diversity in Three  Metropolitan Areas	NA	NA	Office Map study	FWD:AGAGTTTGATCCTGGCTCAG; REV:TGCTGCCTCCCGTAGGAGT	22666400	NA	NA	NA	Dan Knights/Justin Kuczinsky	CCME	GAZ:United States of America	NA	NA	0	NA	NA	NA	NA	n	NA	Scott Kelley	Kelley_office_contamination	pyrosequencing	NA	NA	NA	metagenomes	NA	ENVO:terrestrial biome	FLX	NA	NA	NA	NA	NA	Kelley_office_contamination	Kelley_Office_Map	408169	CCME	NA	NA	NA	1479	FLX protocol	NA	NA	n	NA	NA	ENVO:surface	16S rRNA	NA	NA	V2	CCME	NA	Phone	NA	NA	NA	40.71435	NA	NA	Swab_sample_from_man_phone_in_New_York_office_room_31
NYMP5.522445	TCACGATTAGCG	CATGCTGCCTCCCGTAGGAGT	NA	NA	NewYork	Phone	NA	NA	Phone	y	2011	Kelley_Office_Map	NA	FWIRNKE01	NA	mimarks-survey	NA	NA	NA	NA	0	NA	NA	9.777	2011	NA	0	NA	NA	male	NA	NA	NA	drscottkelley@gmail.com	NA	5	NA	".1,g"	NA	NA	"Samples from office surfaces in New York, San Francisco and Tucson"	NA	NA	-74.00597	y	NA	CCME	NA	NA	NA	"People in developed countries spend approximately 90% of their lives indoors, yet we know little about the source and diversity of microbes in built environments. In this study, we combined culture-based cell counting and multiplexed pyrosequencing of environmental ribosomal RNA (rRNA) gene sequences to investigate office space bacterial diversity in three metropolitan areas. Five surfaces common to all offices were sampled using sterile double-tipped swabs, one tip for culturing and one for DNA extraction, in 30 different offices per city (90 offices, 450 total samples). 16S rRNA gene sequences were PCR amplified using bar-coded universal bacterial primers from 54 of the surfaces (18 per city) and pooled for pyrosequencing. A three-factorial Analysis of Variance (ANOVA) found significant differences in viable bacterial abundance between offices inhabited by men or women, among the various surface types, and among cities. Multiplex pyrosequencing identified more than 500 bacterial genera from 20 different bacterial divisions. The most abundant of these genera tended to be common inhabitants of human skin, nasal, oral or intestinal cavities. Other commonly occurring genera appeared to have environmental origins (e.g., soils). There were no significant differences in the bacterial diversity between offices inhabited by men or women or among surfaces, but the bacterial community diversity of the Tucson samples was clearly distinguishable from that of New York and San Francisco, which were indistinguishable. Overall, our comprehensive molecular analysis of office building microbial diversity shows the potential of these methods for studying patterns and origins of indoor bacterial contamination. Humans move through a sea of microbial life that is seldom perceived except in the context of potential disease and decay Feazel et al. (2009)."	ENVO:surface	NA	Office Space Bacterial Abundance and Diversity in Three  Metropolitan Areas	NA	NA	Office Map study	FWD:AGAGTTTGATCCTGGCTCAG; REV:TGCTGCCTCCCGTAGGAGT	22666400	NA	NA	NA	Dan Knights/Justin Kuczinsky	CCME	GAZ:United States of America	NA	NA	0	NA	NA	NA	NA	n	NA	Scott Kelley	Kelley_office_contamination	pyrosequencing	NA	NA	NA	metagenomes	NA	ENVO:terrestrial biome	FLX	NA	NA	NA	NA	NA	Kelley_office_contamination	Kelley_Office_Map	408169	CCME	NA	NA	NA	1479	FLX protocol	NA	NA	n	NA	NA	ENVO:surface	16S rRNA	NA	NA	V2	CCME	NA	Phone	NA	NA	NA	40.71435	NA	NA	Swab_sample_from_man_phone_in_New_York_office_room_5
SFFP10.522484	CGAGGCTCAGTA	CATGCTGCCTCCCGTAGGAGT	NA	NA	SanFrancisco	Phone	NA	NA	Phone	y	2011	Kelley_Office_Map	NA	FWIRNKE01	NA	mimarks-survey	NA	NA	NA	NA	0	NA	NA	15.5624	2011	NA	0	NA	NA	female	NA	NA	NA	drscottkelley@gmail.com	NA	10	NA	".1,g"	NA	NA	"Samples from office surfaces in New York, San Francisco and Tucson"	NA	NA	-122.41942	y	NA	CCME	NA	NA	NA	"People in developed countries spend approximately 90% of their lives indoors, yet we know little about the source and diversity of microbes in built environments. In this study, we combined culture-based cell counting and multiplexed pyrosequencing of environmental ribosomal RNA (rRNA) gene sequences to investigate office space bacterial diversity in three metropolitan areas. Five surfaces common to all offices were sampled using sterile double-tipped swabs, one tip for culturing and one for DNA extraction, in 30 different offices per city (90 offices, 450 total samples). 16S rRNA gene sequences were PCR amplified using bar-coded universal bacterial primers from 54 of the surfaces (18 per city) and pooled for pyrosequencing. A three-factorial Analysis of Variance (ANOVA) found significant differences in viable bacterial abundance between offices inhabited by men or women, among the various surface types, and among cities. Multiplex pyrosequencing identified more than 500 bacterial genera from 20 different bacterial divisions. The most abundant of these genera tended to be common inhabitants of human skin, nasal, oral or intestinal cavities. Other commonly occurring genera appeared to have environmental origins (e.g., soils). There were no significant differences in the bacterial diversity between offices inhabited by men or women or among surfaces, but the bacterial community diversity of the Tucson samples was clearly distinguishable from that of New York and San Francisco, which were indistinguishable. Overall, our comprehensive molecular analysis of office building microbial diversity shows the potential of these methods for studying patterns and origins of indoor bacterial contamination. Humans move through a sea of microbial life that is seldom perceived except in the context of potential disease and decay Feazel et al. (2009)."	ENVO:surface	NA	Office Space Bacterial Abundance and Diversity in Three  Metropolitan Areas	NA	NA	Office Map study	FWD:AGAGTTTGATCCTGGCTCAG; REV:TGCTGCCTCCCGTAGGAGT	22666400	NA	NA	NA	Dan Knights/Justin Kuczinsky	CCME	GAZ:United States of America	NA	NA	0	NA	NA	NA	NA	n	NA	Scott Kelley	Kelley_office_contamination	pyrosequencing	NA	NA	NA	metagenomes	NA	ENVO:terrestrial biome	FLX	NA	NA	NA	NA	NA	Kelley_office_contamination	Kelley_Office_Map	408169	CCME	NA	NA	NA	1479	FLX protocol	NA	NA	n	NA	NA	ENVO:surface	16S rRNA	NA	NA	V2	CCME	NA	Phone	NA	NA	NA	37.77493	NA	NA	Swab_sample_from_woman_phone_in_San_Francisco_office_room_10
SFFP23.522478	CGAGTCTAGTTG	CATGCTGCCTCCCGTAGGAGT	NA	NA	SanFrancisco	Phone	NA	NA	Phone	y	2011	Kelley_Office_Map	NA	FWIRNKE01	NA	mimarks-survey	NA	NA	NA	NA	0	NA	NA	15.5624	2011	NA	0	NA	NA	female	NA	NA	NA	drscottkelley@gmail.com	NA	23	NA	".1,g"	NA	NA	"Samples from office surfaces in New York, San Francisco and Tucson"	NA	NA	-122.41942	y	NA	CCME	NA	NA	NA	"People in developed countries spend approximately 90% of their lives indoors, yet we know little about the source and diversity of microbes in built environments. In this study, we combined culture-based cell counting and multiplexed pyrosequencing of environmental ribosomal RNA (rRNA) gene sequences to investigate office space bacterial diversity in three metropolitan areas. Five surfaces common to all offices were sampled using sterile double-tipped swabs, one tip for culturing and one for DNA extraction, in 30 different offices per city (90 offices, 450 total samples). 16S rRNA gene sequences were PCR amplified using bar-coded universal bacterial primers from 54 of the surfaces (18 per city) and pooled for pyrosequencing. A three-factorial Analysis of Variance (ANOVA) found significant differences in viable bacterial abundance between offices inhabited by men or women, among the various surface types, and among cities. Multiplex pyrosequencing identified more than 500 bacterial genera from 20 different bacterial divisions. The most abundant of these genera tended to be common inhabitants of human skin, nasal, oral or intestinal cavities. Other commonly occurring genera appeared to have environmental origins (e.g., soils). There were no significant differences in the bacterial diversity between offices inhabited by men or women or among surfaces, but the bacterial community diversity of the Tucson samples was clearly distinguishable from that of New York and San Francisco, which were indistinguishable. Overall, our comprehensive molecular analysis of office building microbial diversity shows the potential of these methods for studying patterns and origins of indoor bacterial contamination. Humans move through a sea of microbial life that is seldom perceived except in the context of potential disease and decay Feazel et al. (2009)."	ENVO:surface	NA	Office Space Bacterial Abundance and Diversity in Three  Metropolitan Areas	NA	NA	Office Map study	FWD:AGAGTTTGATCCTGGCTCAG; REV:TGCTGCCTCCCGTAGGAGT	22666400	NA	NA	NA	Dan Knights/Justin Kuczinsky	CCME	GAZ:United States of America	NA	NA	0	NA	NA	NA	NA	n	NA	Scott Kelley	Kelley_office_contamination	pyrosequencing	NA	NA	NA	metagenomes	NA	ENVO:terrestrial biome	FLX	NA	NA	NA	NA	NA	Kelley_office_contamination	Kelley_Office_Map	408169	CCME	NA	NA	NA	1479	FLX protocol	NA	NA	n	NA	NA	ENVO:surface	16S rRNA	NA	NA	V2	CCME	NA	Phone	NA	NA	NA	37.77493	NA	NA	Swab_sample_from_woman_phone_in_San_Francisco_office_room_23
SFFP3.522454	CGAGCAGCACAT	CATGCTGCCTCCCGTAGGAGT	NA	NA	SanFrancisco	Phone	NA	NA	Phone	y	2011	Kelley_Office_Map	NA	FWIRNKE01	NA	mimarks-survey	NA	NA	NA	NA	0	NA	NA	15.5624	2011	NA	0	NA	NA	female	NA	NA	NA	drscottkelley@gmail.com	NA	3	NA	".1,g"	NA	NA	"Samples from office surfaces in New York, San Francisco and Tucson"	NA	NA	-122.41942	y	NA	CCME	NA	NA	NA	"People in developed countries spend approximately 90% of their lives indoors, yet we know little about the source and diversity of microbes in built environments. In this study, we combined culture-based cell counting and multiplexed pyrosequencing of environmental ribosomal RNA (rRNA) gene sequences to investigate office space bacterial diversity in three metropolitan areas. Five surfaces common to all offices were sampled using sterile double-tipped swabs, one tip for culturing and one for DNA extraction, in 30 different offices per city (90 offices, 450 total samples). 16S rRNA gene sequences were PCR amplified using bar-coded universal bacterial primers from 54 of the surfaces (18 per city) and pooled for pyrosequencing. A three-factorial Analysis of Variance (ANOVA) found significant differences in viable bacterial abundance between offices inhabited by men or women, among the various surface types, and among cities. Multiplex pyrosequencing identified more than 500 bacterial genera from 20 different bacterial divisions. The most abundant of these genera tended to be common inhabitants of human skin, nasal, oral or intestinal cavities. Other commonly occurring genera appeared to have environmental origins (e.g., soils). There were no significant differences in the bacterial diversity between offices inhabited by men or women or among surfaces, but the bacterial community diversity of the Tucson samples was clearly distinguishable from that of New York and San Francisco, which were indistinguishable. Overall, our comprehensive molecular analysis of office building microbial diversity shows the potential of these methods for studying patterns and origins of indoor bacterial contamination. Humans move through a sea of microbial life that is seldom perceived except in the context of potential disease and decay Feazel et al. (2009)."	ENVO:surface	NA	Office Space Bacterial Abundance and Diversity in Three  Metropolitan Areas	NA	NA	Office Map study	FWD:AGAGTTTGATCCTGGCTCAG; REV:TGCTGCCTCCCGTAGGAGT	22666400	NA	NA	NA	Dan Knights/Justin Kuczinsky	CCME	GAZ:United States of America	NA	NA	0	NA	NA	NA	NA	n	NA	Scott Kelley	Kelley_office_contamination	pyrosequencing	NA	NA	NA	metagenomes	NA	ENVO:terrestrial biome	FLX	NA	NA	NA	NA	NA	Kelley_office_contamination	Kelley_Office_Map	408169	CCME	NA	NA	NA	1479	FLX protocol	NA	NA	n	NA	NA	ENVO:surface	16S rRNA	NA	NA	V2	CCME	NA	Phone	NA	NA	NA	37.77493	NA	NA	Swab_sample_from_woman_phone_in_San_Francisco_office_room_3
SFMP16.522460	TCCTGAGATACG	CATGCTGCCTCCCGTAGGAGT	NA	NA	SanFrancisco	Phone	NA	NA	Phone	y	2011	Kelley_Office_Map	NA	FWIRNKE01	NA	mimarks-survey	NA	NA	NA	NA	0	NA	NA	15.5624	2011	NA	0	NA	NA	male	NA	NA	NA	drscottkelley@gmail.com	NA	16	NA	".1,g"	NA	NA	"Samples from office surfaces in New York, San Francisco and Tucson"	NA	NA	-122.41942	y	NA	CCME	NA	NA	NA	"People in developed countries spend approximately 90% of their lives indoors, yet we know little about the source and diversity of microbes in built environments. In this study, we combined culture-based cell counting and multiplexed pyrosequencing of environmental ribosomal RNA (rRNA) gene sequences to investigate office space bacterial diversity in three metropolitan areas. Five surfaces common to all offices were sampled using sterile double-tipped swabs, one tip for culturing and one for DNA extraction, in 30 different offices per city (90 offices, 450 total samples). 16S rRNA gene sequences were PCR amplified using bar-coded universal bacterial primers from 54 of the surfaces (18 per city) and pooled for pyrosequencing. A three-factorial Analysis of Variance (ANOVA) found significant differences in viable bacterial abundance between offices inhabited by men or women, among the various surface types, and among cities. Multiplex pyrosequencing identified more than 500 bacterial genera from 20 different bacterial divisions. The most abundant of these genera tended to be common inhabitants of human skin, nasal, oral or intestinal cavities. Other commonly occurring genera appeared to have environmental origins (e.g., soils). There were no significant differences in the bacterial diversity between offices inhabited by men or women or among surfaces, but the bacterial community diversity of the Tucson samples was clearly distinguishable from that of New York and San Francisco, which were indistinguishable. Overall, our comprehensive molecular analysis of office building microbial diversity shows the potential of these methods for studying patterns and origins of indoor bacterial contamination. Humans move through a sea of microbial life that is seldom perceived except in the context of potential disease and decay Feazel et al. (2009)."	ENVO:surface	NA	Office Space Bacterial Abundance and Diversity in Three  Metropolitan Areas	NA	NA	Office Map study	FWD:AGAGTTTGATCCTGGCTCAG; REV:TGCTGCCTCCCGTAGGAGT	22666400	NA	NA	NA	Dan Knights/Justin Kuczinsky	CCME	GAZ:United States of America	NA	NA	0	NA	NA	NA	NA	n	NA	Scott Kelley	Kelley_office_contamination	pyrosequencing	NA	NA	NA	metagenomes	NA	ENVO:terrestrial biome	FLX	NA	NA	NA	NA	NA	Kelley_office_contamination	Kelley_Office_Map	408169	CCME	NA	NA	NA	1479	FLX protocol	NA	NA	n	NA	NA	ENVO:surface	16S rRNA	NA	NA	V2	CCME	NA	Phone	NA	NA	NA	37.77493	NA	NA	Swab_sample_from_man_phone_in_San_Francisco_office_room_16
SFMP29.522452	TCGACTCCTCGT	CATGCTGCCTCCCGTAGGAGT	NA	NA	SanFrancisco	Phone	NA	NA	Phone	y	2011	Kelley_Office_Map	NA	FWIRNKE01	NA	mimarks-survey	NA	NA	NA	NA	0	NA	NA	15.5624	2011	NA	0	NA	NA	male	NA	NA	NA	drscottkelley@gmail.com	NA	29	NA	".1,g"	NA	NA	"Samples from office surfaces in New York, San Francisco and Tucson"	NA	NA	-122.41942	y	NA	CCME	NA	NA	NA	"People in developed countries spend approximately 90% of their lives indoors, yet we know little about the source and diversity of microbes in built environments. In this study, we combined culture-based cell counting and multiplexed pyrosequencing of environmental ribosomal RNA (rRNA) gene sequences to investigate office space bacterial diversity in three metropolitan areas. Five surfaces common to all offices were sampled using sterile double-tipped swabs, one tip for culturing and one for DNA extraction, in 30 different offices per city (90 offices, 450 total samples). 16S rRNA gene sequences were PCR amplified using bar-coded universal bacterial primers from 54 of the surfaces (18 per city) and pooled for pyrosequencing. A three-factorial Analysis of Variance (ANOVA) found significant differences in viable bacterial abundance between offices inhabited by men or women, among the various surface types, and among cities. Multiplex pyrosequencing identified more than 500 bacterial genera from 20 different bacterial divisions. The most abundant of these genera tended to be common inhabitants of human skin, nasal, oral or intestinal cavities. Other commonly occurring genera appeared to have environmental origins (e.g., soils). There were no significant differences in the bacterial diversity between offices inhabited by men or women or among surfaces, but the bacterial community diversity of the Tucson samples was clearly distinguishable from that of New York and San Francisco, which were indistinguishable. Overall, our comprehensive molecular analysis of office building microbial diversity shows the potential of these methods for studying patterns and origins of indoor bacterial contamination. Humans move through a sea of microbial life that is seldom perceived except in the context of potential disease and decay Feazel et al. (2009)."	ENVO:surface	NA	Office Space Bacterial Abundance and Diversity in Three  Metropolitan Areas	NA	NA	Office Map study	FWD:AGAGTTTGATCCTGGCTCAG; REV:TGCTGCCTCCCGTAGGAGT	22666400	NA	NA	NA	Dan Knights/Justin Kuczinsky	CCME	GAZ:United States of America	NA	NA	0	NA	NA	NA	NA	n	NA	Scott Kelley	Kelley_office_contamination	pyrosequencing	NA	NA	NA	metagenomes	NA	ENVO:terrestrial biome	FLX	NA	NA	NA	NA	NA	Kelley_office_contamination	Kelley_Office_Map	408169	CCME	NA	NA	NA	1479	FLX protocol	NA	NA	n	NA	NA	ENVO:surface	16S rRNA	NA	NA	V2	CCME	NA	Phone	NA	NA	NA	37.77493	NA	NA	Swab_sample_from_man_phone_in_San_Francisco_office_room_29
SFMP7.522455	TCCTCTGTCGAC	CATGCTGCCTCCCGTAGGAGT	NA	NA	SanFrancisco	Phone	NA	NA	Phone	y	2011	Kelley_Office_Map	NA	FWIRNKE01	NA	mimarks-survey	NA	NA	NA	NA	0	NA	NA	15.5624	2011	NA	0	NA	NA	male	NA	NA	NA	drscottkelley@gmail.com	NA	7	NA	".1,g"	NA	NA	"Samples from office surfaces in New York, San Francisco and Tucson"	NA	NA	-122.41942	y	NA	CCME	NA	NA	NA	"People in developed countries spend approximately 90% of their lives indoors, yet we know little about the source and diversity of microbes in built environments. In this study, we combined culture-based cell counting and multiplexed pyrosequencing of environmental ribosomal RNA (rRNA) gene sequences to investigate office space bacterial diversity in three metropolitan areas. Five surfaces common to all offices were sampled using sterile double-tipped swabs, one tip for culturing and one for DNA extraction, in 30 different offices per city (90 offices, 450 total samples). 16S rRNA gene sequences were PCR amplified using bar-coded universal bacterial primers from 54 of the surfaces (18 per city) and pooled for pyrosequencing. A three-factorial Analysis of Variance (ANOVA) found significant differences in viable bacterial abundance between offices inhabited by men or women, among the various surface types, and among cities. Multiplex pyrosequencing identified more than 500 bacterial genera from 20 different bacterial divisions. The most abundant of these genera tended to be common inhabitants of human skin, nasal, oral or intestinal cavities. Other commonly occurring genera appeared to have environmental origins (e.g., soils). There were no significant differences in the bacterial diversity between offices inhabited by men or women or among surfaces, but the bacterial community diversity of the Tucson samples was clearly distinguishable from that of New York and San Francisco, which were indistinguishable. Overall, our comprehensive molecular analysis of office building microbial diversity shows the potential of these methods for studying patterns and origins of indoor bacterial contamination. Humans move through a sea of microbial life that is seldom perceived except in the context of potential disease and decay Feazel et al. (2009)."	ENVO:surface	NA	Office Space Bacterial Abundance and Diversity in Three  Metropolitan Areas	NA	NA	Office Map study	FWD:AGAGTTTGATCCTGGCTCAG; REV:TGCTGCCTCCCGTAGGAGT	22666400	NA	NA	NA	Dan Knights/Justin Kuczinsky	CCME	GAZ:United States of America	NA	NA	0	NA	NA	NA	NA	n	NA	Scott Kelley	Kelley_office_contamination	pyrosequencing	NA	NA	NA	metagenomes	NA	ENVO:terrestrial biome	FLX	NA	NA	NA	NA	NA	Kelley_office_contamination	Kelley_Office_Map	408169	CCME	NA	NA	NA	1479	FLX protocol	NA	NA	n	NA	NA	ENVO:surface	16S rRNA	NA	NA	V2	CCME	NA	Phone	NA	NA	NA	37.77493	NA	NA	Swab_sample_from_man_phone_in_San_Francisco_office_room_7
TUFP22.522474	CGTAAGTCTACT	CATGCTGCCTCCCGTAGGAGT	NA	NA	Tucson	Phone	NA	NA	Phone	y	2011	Kelley_Office_Map	NA	FWIRNKE01	NA	mimarks-survey	NA	NA	NA	NA	0	NA	NA	758.65747	2011	NA	0	NA	NA	female	NA	NA	NA	drscottkelley@gmail.com	NA	22	NA	".1,g"	NA	NA	"Samples from office surfaces in New York, San Francisco and Tucson"	NA	NA	-110.9264	y	NA	CCME	NA	NA	NA	"People in developed countries spend approximately 90% of their lives indoors, yet we know little about the source and diversity of microbes in built environments. In this study, we combined culture-based cell counting and multiplexed pyrosequencing of environmental ribosomal RNA (rRNA) gene sequences to investigate office space bacterial diversity in three metropolitan areas. Five surfaces common to all offices were sampled using sterile double-tipped swabs, one tip for culturing and one for DNA extraction, in 30 different offices per city (90 offices, 450 total samples). 16S rRNA gene sequences were PCR amplified using bar-coded universal bacterial primers from 54 of the surfaces (18 per city) and pooled for pyrosequencing. A three-factorial Analysis of Variance (ANOVA) found significant differences in viable bacterial abundance between offices inhabited by men or women, among the various surface types, and among cities. Multiplex pyrosequencing identified more than 500 bacterial genera from 20 different bacterial divisions. The most abundant of these genera tended to be common inhabitants of human skin, nasal, oral or intestinal cavities. Other commonly occurring genera appeared to have environmental origins (e.g., soils). There were no significant differences in the bacterial diversity between offices inhabited by men or women or among surfaces, but the bacterial community diversity of the Tucson samples was clearly distinguishable from that of New York and San Francisco, which were indistinguishable. Overall, our comprehensive molecular analysis of office building microbial diversity shows the potential of these methods for studying patterns and origins of indoor bacterial contamination. Humans move through a sea of microbial life that is seldom perceived except in the context of potential disease and decay Feazel et al. (2009)."	ENVO:surface	NA	Office Space Bacterial Abundance and Diversity in Three  Metropolitan Areas	NA	NA	Office Map study	FWD:AGAGTTTGATCCTGGCTCAG; REV:TGCTGCCTCCCGTAGGAGT	22666400	NA	NA	NA	Dan Knights/Justin Kuczinsky	CCME	GAZ:United States of America	NA	NA	0	NA	NA	NA	NA	n	NA	Scott Kelley	Kelley_office_contamination	pyrosequencing	NA	NA	NA	metagenomes	NA	ENVO:terrestrial biome	FLX	NA	NA	NA	NA	NA	Kelley_office_contamination	Kelley_Office_Map	408169	CCME	NA	NA	NA	1479	FLX protocol	NA	NA	n	NA	NA	ENVO:surface	16S rRNA	NA	NA	V2	CCME	NA	Phone	NA	NA	NA	32.22174	NA	NA	Swab_sample_from_woman_phone_in_Tucson_office_room_22
TUFP24.522463	CGGCGATGTACA	CATGCTGCCTCCCGTAGGAGT	NA	NA	Tucson	Phone	NA	NA	Phone	y	2011	Kelley_Office_Map	NA	FWIRNKE01	NA	mimarks-survey	NA	NA	NA	NA	0	NA	NA	758.65747	2011	NA	0	NA	NA	female	NA	NA	NA	drscottkelley@gmail.com	NA	24	NA	".1,g"	NA	NA	"Samples from office surfaces in New York, San Francisco and Tucson"	NA	NA	-110.9264	y	NA	CCME	NA	NA	NA	"People in developed countries spend approximately 90% of their lives indoors, yet we know little about the source and diversity of microbes in built environments. In this study, we combined culture-based cell counting and multiplexed pyrosequencing of environmental ribosomal RNA (rRNA) gene sequences to investigate office space bacterial diversity in three metropolitan areas. Five surfaces common to all offices were sampled using sterile double-tipped swabs, one tip for culturing and one for DNA extraction, in 30 different offices per city (90 offices, 450 total samples). 16S rRNA gene sequences were PCR amplified using bar-coded universal bacterial primers from 54 of the surfaces (18 per city) and pooled for pyrosequencing. A three-factorial Analysis of Variance (ANOVA) found significant differences in viable bacterial abundance between offices inhabited by men or women, among the various surface types, and among cities. Multiplex pyrosequencing identified more than 500 bacterial genera from 20 different bacterial divisions. The most abundant of these genera tended to be common inhabitants of human skin, nasal, oral or intestinal cavities. Other commonly occurring genera appeared to have environmental origins (e.g., soils). There were no significant differences in the bacterial diversity between offices inhabited by men or women or among surfaces, but the bacterial community diversity of the Tucson samples was clearly distinguishable from that of New York and San Francisco, which were indistinguishable. Overall, our comprehensive molecular analysis of office building microbial diversity shows the potential of these methods for studying patterns and origins of indoor bacterial contamination. Humans move through a sea of microbial life that is seldom perceived except in the context of potential disease and decay Feazel et al. (2009)."	ENVO:surface	NA	Office Space Bacterial Abundance and Diversity in Three  Metropolitan Areas	NA	NA	Office Map study	FWD:AGAGTTTGATCCTGGCTCAG; REV:TGCTGCCTCCCGTAGGAGT	22666400	NA	NA	NA	Dan Knights/Justin Kuczinsky	CCME	GAZ:United States of America	NA	NA	0	NA	NA	NA	NA	n	NA	Scott Kelley	Kelley_office_contamination	pyrosequencing	NA	NA	NA	metagenomes	NA	ENVO:terrestrial biome	FLX	NA	NA	NA	NA	NA	Kelley_office_contamination	Kelley_Office_Map	408169	CCME	NA	NA	NA	1479	FLX protocol	NA	NA	n	NA	NA	ENVO:surface	16S rRNA	NA	NA	V2	CCME	NA	Phone	NA	NA	NA	32.22174	NA	NA	Swab_sample_from_woman_phone_in_Tucson_office_room_24
TUFP4.522432	CGTACAGTTATC	CATGCTGCCTCCCGTAGGAGT	NA	NA	Tucson	Phone	NA	NA	Phone	y	2011	Kelley_Office_Map	NA	FWIRNKE01	NA	mimarks-survey	NA	NA	NA	NA	0	NA	NA	758.65747	2011	NA	0	NA	NA	female	NA	NA	NA	drscottkelley@gmail.com	NA	4	NA	".1,g"	NA	NA	"Samples from office surfaces in New York, San Francisco and Tucson"	NA	NA	-110.9264	y	NA	CCME	NA	NA	NA	"People in developed countries spend approximately 90% of their lives indoors, yet we know little about the source and diversity of microbes in built environments. In this study, we combined culture-based cell counting and multiplexed pyrosequencing of environmental ribosomal RNA (rRNA) gene sequences to investigate office space bacterial diversity in three metropolitan areas. Five surfaces common to all offices were sampled using sterile double-tipped swabs, one tip for culturing and one for DNA extraction, in 30 different offices per city (90 offices, 450 total samples). 16S rRNA gene sequences were PCR amplified using bar-coded universal bacterial primers from 54 of the surfaces (18 per city) and pooled for pyrosequencing. A three-factorial Analysis of Variance (ANOVA) found significant differences in viable bacterial abundance between offices inhabited by men or women, among the various surface types, and among cities. Multiplex pyrosequencing identified more than 500 bacterial genera from 20 different bacterial divisions. The most abundant of these genera tended to be common inhabitants of human skin, nasal, oral or intestinal cavities. Other commonly occurring genera appeared to have environmental origins (e.g., soils). There were no significant differences in the bacterial diversity between offices inhabited by men or women or among surfaces, but the bacterial community diversity of the Tucson samples was clearly distinguishable from that of New York and San Francisco, which were indistinguishable. Overall, our comprehensive molecular analysis of office building microbial diversity shows the potential of these methods for studying patterns and origins of indoor bacterial contamination. Humans move through a sea of microbial life that is seldom perceived except in the context of potential disease and decay Feazel et al. (2009)."	ENVO:surface	NA	Office Space Bacterial Abundance and Diversity in Three  Metropolitan Areas	NA	NA	Office Map study	FWD:AGAGTTTGATCCTGGCTCAG; REV:TGCTGCCTCCCGTAGGAGT	22666400	NA	NA	NA	Dan Knights/Justin Kuczinsky	CCME	GAZ:United States of America	NA	NA	0	NA	NA	NA	NA	n	NA	Scott Kelley	Kelley_office_contamination	pyrosequencing	NA	NA	NA	metagenomes	NA	ENVO:terrestrial biome	FLX	NA	NA	NA	NA	NA	Kelley_office_contamination	Kelley_Office_Map	408169	CCME	NA	NA	NA	1479	FLX protocol	NA	NA	n	NA	NA	ENVO:surface	16S rRNA	NA	NA	V2	CCME	NA	Phone	NA	NA	NA	32.22174	NA	NA	Swab_sample_from_woman_phone_in_Tucson_office_room_4
TUMP15.522439	CGCACTCTAGAA	CATGCTGCCTCCCGTAGGAGT	NA	NA	Tucson	Phone	NA	NA	Phone	y	2011	Kelley_Office_Map	NA	FWIRNKE01	NA	mimarks-survey	NA	NA	NA	NA	0	NA	NA	758.65747	2011	NA	0	NA	NA	male	NA	NA	NA	drscottkelley@gmail.com	NA	15	NA	".1,g"	NA	NA	"Samples from office surfaces in New York, San Francisco and Tucson"	NA	NA	-110.9264	y	NA	CCME	NA	NA	NA	"People in developed countries spend approximately 90% of their lives indoors, yet we know little about the source and diversity of microbes in built environments. In this study, we combined culture-based cell counting and multiplexed pyrosequencing of environmental ribosomal RNA (rRNA) gene sequences to investigate office space bacterial diversity in three metropolitan areas. Five surfaces common to all offices were sampled using sterile double-tipped swabs, one tip for culturing and one for DNA extraction, in 30 different offices per city (90 offices, 450 total samples). 16S rRNA gene sequences were PCR amplified using bar-coded universal bacterial primers from 54 of the surfaces (18 per city) and pooled for pyrosequencing. A three-factorial Analysis of Variance (ANOVA) found significant differences in viable bacterial abundance between offices inhabited by men or women, among the various surface types, and among cities. Multiplex pyrosequencing identified more than 500 bacterial genera from 20 different bacterial divisions. The most abundant of these genera tended to be common inhabitants of human skin, nasal, oral or intestinal cavities. Other commonly occurring genera appeared to have environmental origins (e.g., soils). There were no significant differences in the bacterial diversity between offices inhabited by men or women or among surfaces, but the bacterial community diversity of the Tucson samples was clearly distinguishable from that of New York and San Francisco, which were indistinguishable. Overall, our comprehensive molecular analysis of office building microbial diversity shows the potential of these methods for studying patterns and origins of indoor bacterial contamination. Humans move through a sea of microbial life that is seldom perceived except in the context of potential disease and decay Feazel et al. (2009)."	ENVO:surface	NA	Office Space Bacterial Abundance and Diversity in Three  Metropolitan Areas	NA	NA	Office Map study	FWD:AGAGTTTGATCCTGGCTCAG; REV:TGCTGCCTCCCGTAGGAGT	22666400	NA	NA	NA	Dan Knights/Justin Kuczinsky	CCME	GAZ:United States of America	NA	NA	0	NA	NA	NA	NA	n	NA	Scott Kelley	Kelley_office_contamination	pyrosequencing	NA	NA	NA	metagenomes	NA	ENVO:terrestrial biome	FLX	NA	NA	NA	NA	NA	Kelley_office_contamination	Kelley_Office_Map	408169	CCME	NA	NA	NA	1479	FLX protocol	NA	NA	n	NA	NA	ENVO:surface	16S rRNA	NA	NA	V2	CCME	NA	Phone	NA	NA	NA	32.22174	NA	NA	Swab_sample_from_man_phone_in_Tucson_office_room_15
TUMP18.522485	CGCAGACAGACT	CATGCTGCCTCCCGTAGGAGT	NA	NA	Tucson	Phone	NA	NA	Phone	y	2011	Kelley_Office_Map	NA	FWIRNKE01	NA	mimarks-survey	NA	NA	NA	NA	0	NA	NA	758.65747	2011	NA	0	NA	NA	male	NA	NA	NA	drscottkelley@gmail.com	NA	18	NA	".1,g"	NA	NA	"Samples from office surfaces in New York, San Francisco and Tucson"	NA	NA	-110.9264	y	NA	CCME	NA	NA	NA	"People in developed countries spend approximately 90% of their lives indoors, yet we know little about the source and diversity of microbes in built environments. In this study, we combined culture-based cell counting and multiplexed pyrosequencing of environmental ribosomal RNA (rRNA) gene sequences to investigate office space bacterial diversity in three metropolitan areas. Five surfaces common to all offices were sampled using sterile double-tipped swabs, one tip for culturing and one for DNA extraction, in 30 different offices per city (90 offices, 450 total samples). 16S rRNA gene sequences were PCR amplified using bar-coded universal bacterial primers from 54 of the surfaces (18 per city) and pooled for pyrosequencing. A three-factorial Analysis of Variance (ANOVA) found significant differences in viable bacterial abundance between offices inhabited by men or women, among the various surface types, and among cities. Multiplex pyrosequencing identified more than 500 bacterial genera from 20 different bacterial divisions. The most abundant of these genera tended to be common inhabitants of human skin, nasal, oral or intestinal cavities. Other commonly occurring genera appeared to have environmental origins (e.g., soils). There were no significant differences in the bacterial diversity between offices inhabited by men or women or among surfaces, but the bacterial community diversity of the Tucson samples was clearly distinguishable from that of New York and San Francisco, which were indistinguishable. Overall, our comprehensive molecular analysis of office building microbial diversity shows the potential of these methods for studying patterns and origins of indoor bacterial contamination. Humans move through a sea of microbial life that is seldom perceived except in the context of potential disease and decay Feazel et al. (2009)."	ENVO:surface	NA	Office Space Bacterial Abundance and Diversity in Three  Metropolitan Areas	NA	NA	Office Map study	FWD:AGAGTTTGATCCTGGCTCAG; REV:TGCTGCCTCCCGTAGGAGT	22666400	NA	NA	NA	Dan Knights/Justin Kuczinsky	CCME	GAZ:United States of America	NA	NA	0	NA	NA	NA	NA	n	NA	Scott Kelley	Kelley_office_contamination	pyrosequencing	NA	NA	NA	metagenomes	NA	ENVO:terrestrial biome	FLX	NA	NA	NA	NA	NA	Kelley_office_contamination	Kelley_Office_Map	408169	CCME	NA	NA	NA	1479	FLX protocol	NA	NA	n	NA	NA	ENVO:surface	16S rRNA	NA	NA	V2	CCME	NA	Phone	NA	NA	NA	32.22174	NA	NA	Swab_sample_from_man_phone_in_Tucson_office_room_18
TUMP23.522457	CGCAGCGGTATA	CATGCTGCCTCCCGTAGGAGT	NA	NA	Tucson	Phone	NA	NA	Phone	y	2011	Kelley_Office_Map	NA	FWIRNKE01	NA	mimarks-survey	NA	NA	NA	NA	0	NA	NA	758.65747	2011	NA	0	NA	NA	male	NA	NA	NA	drscottkelley@gmail.com	NA	23	NA	".1,g"	NA	NA	"Samples from office surfaces in New York, San Francisco and Tucson"	NA	NA	-110.9264	y	NA	CCME	NA	NA	NA	"People in developed countries spend approximately 90% of their lives indoors, yet we know little about the source and diversity of microbes in built environments. In this study, we combined culture-based cell counting and multiplexed pyrosequencing of environmental ribosomal RNA (rRNA) gene sequences to investigate office space bacterial diversity in three metropolitan areas. Five surfaces common to all offices were sampled using sterile double-tipped swabs, one tip for culturing and one for DNA extraction, in 30 different offices per city (90 offices, 450 total samples). 16S rRNA gene sequences were PCR amplified using bar-coded universal bacterial primers from 54 of the surfaces (18 per city) and pooled for pyrosequencing. A three-factorial Analysis of Variance (ANOVA) found significant differences in viable bacterial abundance between offices inhabited by men or women, among the various surface types, and among cities. Multiplex pyrosequencing identified more than 500 bacterial genera from 20 different bacterial divisions. The most abundant of these genera tended to be common inhabitants of human skin, nasal, oral or intestinal cavities. Other commonly occurring genera appeared to have environmental origins (e.g., soils). There were no significant differences in the bacterial diversity between offices inhabited by men or women or among surfaces, but the bacterial community diversity of the Tucson samples was clearly distinguishable from that of New York and San Francisco, which were indistinguishable. Overall, our comprehensive molecular analysis of office building microbial diversity shows the potential of these methods for studying patterns and origins of indoor bacterial contamination. Humans move through a sea of microbial life that is seldom perceived except in the context of potential disease and decay Feazel et al. (2009)."	ENVO:surface	NA	Office Space Bacterial Abundance and Diversity in Three  Metropolitan Areas	NA	NA	Office Map study	FWD:AGAGTTTGATCCTGGCTCAG; REV:TGCTGCCTCCCGTAGGAGT	22666400	NA	NA	NA	Dan Knights/Justin Kuczinsky	CCME	GAZ:United States of America	NA	NA	0	NA	NA	NA	NA	n	NA	Scott Kelley	Kelley_office_contamination	pyrosequencing	NA	NA	NA	metagenomes	NA	ENVO:terrestrial biome	FLX	NA	NA	NA	NA	NA	Kelley_office_contamination	Kelley_Office_Map	408169	CCME	NA	NA	NA	1479	FLX protocol	NA	NA	n	NA	NA	ENVO:surface	16S rRNA	NA	NA	V2	CCME	NA	Phone	NA	NA	NA	32.22174	NA	NA	Swab_sample_from_man_phone_in_Tucson_office_room_23
TUMP25.522443	CGCATGAGGATC	CATGCTGCCTCCCGTAGGAGT	NA	NA	Tucson	Phone	NA	NA	Phone	y	2011	Kelley_Office_Map	NA	FWIRNKE01	NA	mimarks-survey	NA	NA	NA	NA	0	NA	NA	758.65747	2011	NA	0	NA	NA	male	NA	NA	NA	drscottkelley@gmail.com	NA	25	NA	".1,g"	NA	NA	"Samples from office surfaces in New York, San Francisco and Tucson"	NA	NA	-110.9264	y	NA	CCME	NA	NA	NA	"People in developed countries spend approximately 90% of their lives indoors, yet we know little about the source and diversity of microbes in built environments. In this study, we combined culture-based cell counting and multiplexed pyrosequencing of environmental ribosomal RNA (rRNA) gene sequences to investigate office space bacterial diversity in three metropolitan areas. Five surfaces common to all offices were sampled using sterile double-tipped swabs, one tip for culturing and one for DNA extraction, in 30 different offices per city (90 offices, 450 total samples). 16S rRNA gene sequences were PCR amplified using bar-coded universal bacterial primers from 54 of the surfaces (18 per city) and pooled for pyrosequencing. A three-factorial Analysis of Variance (ANOVA) found significant differences in viable bacterial abundance between offices inhabited by men or women, among the various surface types, and among cities. Multiplex pyrosequencing identified more than 500 bacterial genera from 20 different bacterial divisions. The most abundant of these genera tended to be common inhabitants of human skin, nasal, oral or intestinal cavities. Other commonly occurring genera appeared to have environmental origins (e.g., soils). There were no significant differences in the bacterial diversity between offices inhabited by men or women or among surfaces, but the bacterial community diversity of the Tucson samples was clearly distinguishable from that of New York and San Francisco, which were indistinguishable. Overall, our comprehensive molecular analysis of office building microbial diversity shows the potential of these methods for studying patterns and origins of indoor bacterial contamination. Humans move through a sea of microbial life that is seldom perceived except in the context of potential disease and decay Feazel et al. (2009)."	ENVO:surface	NA	Office Space Bacterial Abundance and Diversity in Three  Metropolitan Areas	NA	NA	Office Map study	FWD:AGAGTTTGATCCTGGCTCAG; REV:TGCTGCCTCCCGTAGGAGT	22666400	NA	NA	NA	Dan Knights/Justin Kuczinsky	CCME	GAZ:United States of America	NA	NA	0	NA	NA	NA	NA	n	NA	Scott Kelley	Kelley_office_contamination	pyrosequencing	NA	NA	NA	metagenomes	NA	ENVO:terrestrial biome	FLX	NA	NA	NA	NA	NA	Kelley_office_contamination	Kelley_Office_Map	408169	CCME	NA	NA	NA	1479	FLX protocol	NA	NA	n	NA	NA	ENVO:surface	16S rRNA	NA	NA	V2	CCME	NA	Phone	NA	NA	NA	32.22174	NA	NA	Swab_sample_from_man_phone_in_Tucson_office_room_25
BB2.141659	AAGCTGCAGTCG	CATGCTGCCTCCCGTAGGAGT	chris.lauber@gmail.com	NA	NA	ENVO:forest_soil	Soil	source	NA	n	5/28/08	Pyrosequencing-Based Assessment of Soil pH as a Predictor of Soil Bacterial Community Structure at the Continental Scale	NA	FA6P1OK	NA	mimarks-survey	NA	NA	NA	52.2	0	NA	"Bear Brook, ME, USA"	400	2008	NA	0	NA	NA	NA	21.4	6.1	NA	NA	NA	BB2	1	"0.5, g"	NA	NA	targeted gene survey	2.223	NA	-68.1	y	0	CCME	NA	NA	NA	"Soils harbor enormously diverse bacterial populations, and soil bacterial communities can vary greatly in composition across space. However, our understanding of the specific changes in soil bacterial community structure that occur across larger spatial scales is limited because most previous work has focused on either surveying a relatively small number of soils in detail or analyzing a larger number of soils with techniques that provide little detail about the phylogenetic structure of the bacterial communities. Here we used a bar-coded pyrosequencing technique to characterize bacterial communities in 88 soils from across North and South America, obtaining an average of 1,501 sequences per soil. We found that overall bacterial community composition, as measured by pairwise UniFrac distances, was significantly correlated with differences in soil pH (r = 0.79), largely driven by changes in the relative abundances of Acidobacteria, Actinobacteria, and Bacteroidetes across the range of soil pHs. In addition, soil pH explains a significant portion of the variability associated with observed changes in the phylogenetic structure within each dominant lineage. The overall phylogenetic diversity of the bacterial communities was also correlated with soil pH (R2 = 0.50), with peak diversity in soils with near-neutral pHs. Together, these results suggest that the structure of soil bacterial communities is predictable, to some degree, across larger spatial scales, and the effect of soil pH on bacterial community composition is evident at even relatively coarse levels of taxonomic resolution."	ENVO:forest soil	TCAG	Pyrosequencing_based_assessment_of_soil_pH_as_a_predictor_of_soil_bacterial_community_structure_at_the_continental_scale	BB2_V2	0	bacterial biogeography	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19502440	NA	NA	NA	Chris Lauber	CCME	GAZ:United States of America	sandy loam	lauber_88_soils	0-0.05	NA	NA	44	NA	n	NA	NA	Lauber_88_soils	pyrosequencing	NA	NA	-680	soil metagenome	NA	ENVO:Temperate broadleaf and mixed forest biome	FLX	BB2_V2	NA	1200	4.6	NA	Lauber88Soils	Pyrosequencing_based_assessment_of_soil_pH_as_a_predictor_of_soil_bacterial_community_structure_at_the_continental_scale	410658	CCME	spodosol	NA	NA	103	19502440	lauber_88_soils	NA	y	NA	BB2_V2	ENVO:soil	16S rRNA	NA	NA	V2	CCME	V2	NA	NA	Engencore	NA	44.86666667	NA	NA	soil metagenome
BF1.141647	AATCAGTCTCGT	CATGCTGCCTCCCGTAGGAGT	chris.lauber@gmail.com	NA	NA	ENVO:forest_soil	Soil	source	NA	n	5/28/08	Pyrosequencing-Based Assessment of Soil pH as a Predictor of Soil Bacterial Community Structure at the Continental Scale	NA	FA6P1OK	NA	mimarks-survey	NA	NA	NA	64.4	0	NA	"Bousson Forest, PA, USA"	390	2008	NA	0	NA	NA	NA	13.7	7.8	NA	NA	NA	BF1	1	"0.5, g"	NA	NA	targeted gene survey	7.019	NA	-80.05	y	0	CCME	NA	NA	NA	"Soils harbor enormously diverse bacterial populations, and soil bacterial communities can vary greatly in composition across space. However, our understanding of the specific changes in soil bacterial community structure that occur across larger spatial scales is limited because most previous work has focused on either surveying a relatively small number of soils in detail or analyzing a larger number of soils with techniques that provide little detail about the phylogenetic structure of the bacterial communities. Here we used a bar-coded pyrosequencing technique to characterize bacterial communities in 88 soils from across North and South America, obtaining an average of 1,501 sequences per soil. We found that overall bacterial community composition, as measured by pairwise UniFrac distances, was significantly correlated with differences in soil pH (r = 0.79), largely driven by changes in the relative abundances of Acidobacteria, Actinobacteria, and Bacteroidetes across the range of soil pHs. In addition, soil pH explains a significant portion of the variability associated with observed changes in the phylogenetic structure within each dominant lineage. The overall phylogenetic diversity of the bacterial communities was also correlated with soil pH (R2 = 0.50), with peak diversity in soils with near-neutral pHs. Together, these results suggest that the structure of soil bacterial communities is predictable, to some degree, across larger spatial scales, and the effect of soil pH on bacterial community composition is evident at even relatively coarse levels of taxonomic resolution."	ENVO:forest soil	TCAG	Pyrosequencing_based_assessment_of_soil_pH_as_a_predictor_of_soil_bacterial_community_structure_at_the_continental_scale	BF1_V2	0	bacterial biogeography	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19502440	NA	NA	NA	Chris Lauber	CCME	GAZ:United States of America	loam	lauber_88_soils	0-0.05	NA	NA	60	NA	n	NA	NA	Lauber_88_soils	pyrosequencing	NA	NA	-451	soil metagenome	NA	ENVO:Temperate broadleaf and mixed forest biome	FLX	BF1_V2	NA	1000	4.05	NA	Lauber88Soils	Pyrosequencing_based_assessment_of_soil_pH_as_a_predictor_of_soil_bacterial_community_structure_at_the_continental_scale	410658	CCME	alfisol	NA	NA	103	19502440	lauber_88_soils	NA	y	NA	BF1_V2	ENVO:soil	16S rRNA	NA	NA	V2	CCME	V2	NA	NA	Engencore	NA	41.58333333	NA	NA	soil metagenome
BF2.141708	AATCGTGACTCG	CATGCTGCCTCCCGTAGGAGT	chris.lauber@gmail.com	NA	NA	ENVO:forest_soil	Soil	source	NA	n	5/28/08	Pyrosequencing-Based Assessment of Soil pH as a Predictor of Soil Bacterial Community Structure at the Continental Scale	NA	FA6P1OK	NA	mimarks-survey	NA	NA	NA	95.2	0	NA	"Bousson Forest, PA, USA"	390	2008	NA	0	NA	NA	NA	15.2	7.8	NA	NA	NA	BF2	1	"0.5, g"	NA	NA	targeted gene survey	8.718	NA	-80.05	y	0	CCME	NA	NA	NA	"Soils harbor enormously diverse bacterial populations, and soil bacterial communities can vary greatly in composition across space. However, our understanding of the specific changes in soil bacterial community structure that occur across larger spatial scales is limited because most previous work has focused on either surveying a relatively small number of soils in detail or analyzing a larger number of soils with techniques that provide little detail about the phylogenetic structure of the bacterial communities. Here we used a bar-coded pyrosequencing technique to characterize bacterial communities in 88 soils from across North and South America, obtaining an average of 1,501 sequences per soil. We found that overall bacterial community composition, as measured by pairwise UniFrac distances, was significantly correlated with differences in soil pH (r = 0.79), largely driven by changes in the relative abundances of Acidobacteria, Actinobacteria, and Bacteroidetes across the range of soil pHs. In addition, soil pH explains a significant portion of the variability associated with observed changes in the phylogenetic structure within each dominant lineage. The overall phylogenetic diversity of the bacterial communities was also correlated with soil pH (R2 = 0.50), with peak diversity in soils with near-neutral pHs. Together, these results suggest that the structure of soil bacterial communities is predictable, to some degree, across larger spatial scales, and the effect of soil pH on bacterial community composition is evident at even relatively coarse levels of taxonomic resolution."	ENVO:forest soil	TCAG	Pyrosequencing_based_assessment_of_soil_pH_as_a_predictor_of_soil_bacterial_community_structure_at_the_continental_scale	BF2_V2	0	bacterial biogeography	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19502440	NA	NA	NA	Chris Lauber	CCME	GAZ:United States of America	loam	lauber_88_soils	0-0.05	NA	NA	56	NA	n	NA	NA	Lauber_88_soils	pyrosequencing	NA	NA	-451	soil metagenome	NA	ENVO:Temperate broadleaf and mixed forest biome	FLX	BF2_V2	NA	1000	3.61	NA	Lauber88Soils	Pyrosequencing_based_assessment_of_soil_pH_as_a_predictor_of_soil_bacterial_community_structure_at_the_continental_scale	410658	CCME	alfisol	NA	NA	103	19502440	lauber_88_soils	NA	y	NA	BF2_V2	ENVO:soil	16S rRNA	NA	NA	V2	CCME	V2	NA	NA	Engencore	NA	41.58333333	NA	NA	soil metagenome
BZ1.141724	ACACATGTCTAC	CATGCTGCCTCCCGTAGGAGT	chris.lauber@gmail.com	NA	NA	ENVO:forest_soil	Soil	source	NA	n	5/28/08	Pyrosequencing-Based Assessment of Soil pH as a Predictor of Soil Bacterial Community Structure at the Continental Scale	NA	FA6P1OK	NA	mimarks-survey	NA	NA	NA	30.3	0	NA	"Bonanza Creek LTER, AK USA"	300	2008	NA	0	NA	NA	NA	19.4	-2.9	NA	NA	NA	BZ1	1	"0.5, g"	NA	NA	targeted gene survey	4.185	NA	-148.25	y	0	CCME	NA	NA	NA	"Soils harbor enormously diverse bacterial populations, and soil bacterial communities can vary greatly in composition across space. However, our understanding of the specific changes in soil bacterial community structure that occur across larger spatial scales is limited because most previous work has focused on either surveying a relatively small number of soils in detail or analyzing a larger number of soils with techniques that provide little detail about the phylogenetic structure of the bacterial communities. Here we used a bar-coded pyrosequencing technique to characterize bacterial communities in 88 soils from across North and South America, obtaining an average of 1,501 sequences per soil. We found that overall bacterial community composition, as measured by pairwise UniFrac distances, was significantly correlated with differences in soil pH (r = 0.79), largely driven by changes in the relative abundances of Acidobacteria, Actinobacteria, and Bacteroidetes across the range of soil pHs. In addition, soil pH explains a significant portion of the variability associated with observed changes in the phylogenetic structure within each dominant lineage. The overall phylogenetic diversity of the bacterial communities was also correlated with soil pH (R2 = 0.50), with peak diversity in soils with near-neutral pHs. Together, these results suggest that the structure of soil bacterial communities is predictable, to some degree, across larger spatial scales, and the effect of soil pH on bacterial community composition is evident at even relatively coarse levels of taxonomic resolution."	ENVO:forest soil	TCAG	Pyrosequencing_based_assessment_of_soil_pH_as_a_predictor_of_soil_bacterial_community_structure_at_the_continental_scale	BZ1_V2	0	bacterial biogeography	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19502440	NA	NA	NA	Chris Lauber	CCME	GAZ:United States of America	silt loam	lauber_88_soils	0-0.05	NA	NA	80	NA	n	NA	NA	Lauber_88_soils	pyrosequencing	NA	NA	133	soil metagenome	NA	ENVO:forest	FLX	BZ1_V2	NA	260	5.12	NA	Lauber88Soils	Pyrosequencing_based_assessment_of_soil_pH_as_a_predictor_of_soil_bacterial_community_structure_at_the_continental_scale	410658	CCME	gelisol	NA	NA	103	19502440	lauber_88_soils	NA	y	NA	BZ1_V2	ENVO:soil	16S rRNA	NA	NA	V2	CCME	V2	NA	NA	Engencore	NA	64.8	NA	NA	soil metagenome
CA1.141704	ACACGAGCCACA	CATGCTGCCTCCCGTAGGAGT	chris.lauber@gmail.com	NA	NA	ENVO:forest_soil	Soil	source	NA	n	5/28/08	Pyrosequencing-Based Assessment of Soil pH as a Predictor of Soil Bacterial Community Structure at the Continental Scale	NA	FA6P1OK	NA	mimarks-survey	NA	NA	NA	16.7	0	NA	"Cedar Mtn. AZ, USA"	2003	2008	NA	0	NA	NA	NA	13	10.3	NA	NA	NA	CA1	1	"0.5, g"	NA	NA	targeted gene survey	2.276	NA	-111.7666667	y	0	CCME	NA	NA	NA	"Soils harbor enormously diverse bacterial populations, and soil bacterial communities can vary greatly in composition across space. However, our understanding of the specific changes in soil bacterial community structure that occur across larger spatial scales is limited because most previous work has focused on either surveying a relatively small number of soils in detail or analyzing a larger number of soils with techniques that provide little detail about the phylogenetic structure of the bacterial communities. Here we used a bar-coded pyrosequencing technique to characterize bacterial communities in 88 soils from across North and South America, obtaining an average of 1,501 sequences per soil. We found that overall bacterial community composition, as measured by pairwise UniFrac distances, was significantly correlated with differences in soil pH (r = 0.79), largely driven by changes in the relative abundances of Acidobacteria, Actinobacteria, and Bacteroidetes across the range of soil pHs. In addition, soil pH explains a significant portion of the variability associated with observed changes in the phylogenetic structure within each dominant lineage. The overall phylogenetic diversity of the bacterial communities was also correlated with soil pH (R2 = 0.50), with peak diversity in soils with near-neutral pHs. Together, these results suggest that the structure of soil bacterial communities is predictable, to some degree, across larger spatial scales, and the effect of soil pH on bacterial community composition is evident at even relatively coarse levels of taxonomic resolution."	ENVO:forest soil	TCAG	Pyrosequencing_based_assessment_of_soil_pH_as_a_predictor_of_soil_bacterial_community_structure_at_the_continental_scale	CA1_V2	0	bacterial biogeography	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19502440	NA	NA	NA	Chris Lauber	CCME	GAZ:United States of America	silt loam	lauber_88_soils	0-0.05	NA	NA	73	NA	n	NA	NA	Lauber_88_soils	pyrosequencing	NA	NA	198	soil metagenome	NA	ENVO:forest	FLX	CA1_V2	NA	400	7.27	NA	Lauber88Soils	Pyrosequencing_based_assessment_of_soil_pH_as_a_predictor_of_soil_bacterial_community_structure_at_the_continental_scale	410658	CCME	alfisol	NA	NA	103	19502440	lauber_88_soils	NA	y	NA	CA1_V2	ENVO:soil	16S rRNA	NA	NA	V2	CCME	V2	NA	NA	Engencore	NA	36.05	NA	NA	soil metagenome
CF1.141675	ACACTGTTCATG	CATGCTGCCTCCCGTAGGAGT	chris.lauber@gmail.com	NA	NA	ENVO:forest_soil	Soil	source	NA	n	5/28/08	Pyrosequencing-Based Assessment of Soil pH as a Predictor of Soil Bacterial Community Structure at the Continental Scale	NA	FA6P1OK	NA	mimarks-survey	NA	NA	NA	25.6	0	NA	"Catskills, NY USA"	800	2008	NA	0	NA	NA	NA	13.6	5.3	NA	NA	NA	CF1	1	"0.5, g"	NA	NA	targeted gene survey	2.266	NA	-74.25833333	y	0	CCME	NA	NA	NA	"Soils harbor enormously diverse bacterial populations, and soil bacterial communities can vary greatly in composition across space. However, our understanding of the specific changes in soil bacterial community structure that occur across larger spatial scales is limited because most previous work has focused on either surveying a relatively small number of soils in detail or analyzing a larger number of soils with techniques that provide little detail about the phylogenetic structure of the bacterial communities. Here we used a bar-coded pyrosequencing technique to characterize bacterial communities in 88 soils from across North and South America, obtaining an average of 1,501 sequences per soil. We found that overall bacterial community composition, as measured by pairwise UniFrac distances, was significantly correlated with differences in soil pH (r = 0.79), largely driven by changes in the relative abundances of Acidobacteria, Actinobacteria, and Bacteroidetes across the range of soil pHs. In addition, soil pH explains a significant portion of the variability associated with observed changes in the phylogenetic structure within each dominant lineage. The overall phylogenetic diversity of the bacterial communities was also correlated with soil pH (R2 = 0.50), with peak diversity in soils with near-neutral pHs. Together, these results suggest that the structure of soil bacterial communities is predictable, to some degree, across larger spatial scales, and the effect of soil pH on bacterial community composition is evident at even relatively coarse levels of taxonomic resolution."	ENVO:forest soil	TCAG	Pyrosequencing_based_assessment_of_soil_pH_as_a_predictor_of_soil_bacterial_community_structure_at_the_continental_scale	CF1_V2	0	bacterial biogeography	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19502440	NA	NA	NA	Chris Lauber	CCME	GAZ:United States of America	loam	lauber_88_soils	0-0.05	NA	NA	49	NA	n	NA	NA	Lauber_88_soils	pyrosequencing	NA	NA	-859	soil metagenome	NA	ENVO:Temperate broadleaf and mixed forest biome	FLX	CF1_V2	NA	1300	3.92	NA	Lauber88Soils	Pyrosequencing_based_assessment_of_soil_pH_as_a_predictor_of_soil_bacterial_community_structure_at_the_continental_scale	410658	CCME	inceptisol	NA	NA	103	19502440	lauber_88_soils	NA	y	NA	CF1_V2	ENVO:soil	16S rRNA	NA	NA	V2	CCME	V2	NA	NA	Engencore	NA	42.15833333	NA	NA	soil metagenome
CF2.141679	ACAGACCACTCA	CATGCTGCCTCCCGTAGGAGT	chris.lauber@gmail.com	NA	NA	ENVO:forest_soil	Soil	source	NA	n	5/28/08	Pyrosequencing-Based Assessment of Soil pH as a Predictor of Soil Bacterial Community Structure at the Continental Scale	NA	FA6P1OK	NA	mimarks-survey	NA	NA	NA	40.6	0	NA	"Catskills, NY USA"	800	2008	NA	0	NA	NA	NA	19.1	5.3	NA	NA	NA	CF2	1	"0.5, g"	NA	NA	targeted gene survey	3	NA	-74.35	y	0	CCME	NA	NA	NA	"Soils harbor enormously diverse bacterial populations, and soil bacterial communities can vary greatly in composition across space. However, our understanding of the specific changes in soil bacterial community structure that occur across larger spatial scales is limited because most previous work has focused on either surveying a relatively small number of soils in detail or analyzing a larger number of soils with techniques that provide little detail about the phylogenetic structure of the bacterial communities. Here we used a bar-coded pyrosequencing technique to characterize bacterial communities in 88 soils from across North and South America, obtaining an average of 1,501 sequences per soil. We found that overall bacterial community composition, as measured by pairwise UniFrac distances, was significantly correlated with differences in soil pH (r = 0.79), largely driven by changes in the relative abundances of Acidobacteria, Actinobacteria, and Bacteroidetes across the range of soil pHs. In addition, soil pH explains a significant portion of the variability associated with observed changes in the phylogenetic structure within each dominant lineage. The overall phylogenetic diversity of the bacterial communities was also correlated with soil pH (R2 = 0.50), with peak diversity in soils with near-neutral pHs. Together, these results suggest that the structure of soil bacterial communities is predictable, to some degree, across larger spatial scales, and the effect of soil pH on bacterial community composition is evident at even relatively coarse levels of taxonomic resolution."	ENVO:forest soil	TCAG	Pyrosequencing_based_assessment_of_soil_pH_as_a_predictor_of_soil_bacterial_community_structure_at_the_continental_scale	CF2_V2	0	bacterial biogeography	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19502440	NA	NA	NA	Chris Lauber	CCME	GAZ:United States of America	sandy loam	lauber_88_soils	0-0.05	NA	NA	28	NA	n	NA	NA	Lauber_88_soils	pyrosequencing	NA	NA	-859	soil metagenome	NA	ENVO:Temperate broadleaf and mixed forest biome	FLX	CF2_V2	NA	1300	3.63	NA	Lauber88Soils	Pyrosequencing_based_assessment_of_soil_pH_as_a_predictor_of_soil_bacterial_community_structure_at_the_continental_scale	410658	CCME	inceptisol	NA	NA	103	19502440	lauber_88_soils	NA	y	NA	CF2_V2	ENVO:soil	16S rRNA	NA	NA	V2	CCME	V2	NA	NA	Engencore	NA	41.93333333	NA	NA	soil metagenome
CF3.141691	ACAGAGTCGGCT	CATGCTGCCTCCCGTAGGAGT	chris.lauber@gmail.com	NA	NA	ENVO:forest_soil	Soil	source	NA	n	5/28/08	Pyrosequencing-Based Assessment of Soil pH as a Predictor of Soil Bacterial Community Structure at the Continental Scale	NA	FA6P1OK	NA	mimarks-survey	NA	NA	NA	43.3	0	NA	"Catskills, NY USA"	800	2008	NA	0	NA	NA	NA	17	5.3	NA	NA	NA	CF3	1	"0.5, g"	NA	NA	targeted gene survey	2.912	NA	-74.1	y	0	CCME	NA	NA	NA	"Soils harbor enormously diverse bacterial populations, and soil bacterial communities can vary greatly in composition across space. However, our understanding of the specific changes in soil bacterial community structure that occur across larger spatial scales is limited because most previous work has focused on either surveying a relatively small number of soils in detail or analyzing a larger number of soils with techniques that provide little detail about the phylogenetic structure of the bacterial communities. Here we used a bar-coded pyrosequencing technique to characterize bacterial communities in 88 soils from across North and South America, obtaining an average of 1,501 sequences per soil. We found that overall bacterial community composition, as measured by pairwise UniFrac distances, was significantly correlated with differences in soil pH (r = 0.79), largely driven by changes in the relative abundances of Acidobacteria, Actinobacteria, and Bacteroidetes across the range of soil pHs. In addition, soil pH explains a significant portion of the variability associated with observed changes in the phylogenetic structure within each dominant lineage. The overall phylogenetic diversity of the bacterial communities was also correlated with soil pH (R2 = 0.50), with peak diversity in soils with near-neutral pHs. Together, these results suggest that the structure of soil bacterial communities is predictable, to some degree, across larger spatial scales, and the effect of soil pH on bacterial community composition is evident at even relatively coarse levels of taxonomic resolution."	ENVO:forest soil	TCAG	Pyrosequencing_based_assessment_of_soil_pH_as_a_predictor_of_soil_bacterial_community_structure_at_the_continental_scale	CF3_V2	0	bacterial biogeography	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19502440	NA	NA	NA	Chris Lauber	CCME	GAZ:United States of America	silt loam	lauber_88_soils	0-0.05	NA	NA	77	NA	n	NA	NA	Lauber_88_soils	pyrosequencing	NA	NA	-859	soil metagenome	NA	ENVO:Temperate broadleaf and mixed forest biome	FLX	CF3_V2	NA	1300	3.56	NA	Lauber88Soils	Pyrosequencing_based_assessment_of_soil_pH_as_a_predictor_of_soil_bacterial_community_structure_at_the_continental_scale	410658	CCME	inceptisol	NA	NA	103	19502440	lauber_88_soils	NA	y	NA	CF3_V2	ENVO:soil	16S rRNA	NA	NA	V2	CCME	V2	NA	NA	Engencore	NA	42.11666667	NA	NA	soil metagenome
CL1.141697	ACAGCAGTGGTC	CATGCTGCCTCCCGTAGGAGT	chris.lauber@gmail.com	NA	NA	ENVO:forest_soil	Soil	source	NA	n	5/28/08	Pyrosequencing-Based Assessment of Soil pH as a Predictor of Soil Bacterial Community Structure at the Continental Scale	NA	FA6P1OK	NA	mimarks-survey	NA	NA	NA	23.3	0	NA	"Calhoun Experimental Forest, SC USA"	150	2008	NA	0	NA	NA	NA	19.6	15.9	NA	NA	NA	CL1	1	"0.5, g"	NA	NA	targeted gene survey	4.642	NA	-81.66666667	y	0	CCME	NA	NA	NA	"Soils harbor enormously diverse bacterial populations, and soil bacterial communities can vary greatly in composition across space. However, our understanding of the specific changes in soil bacterial community structure that occur across larger spatial scales is limited because most previous work has focused on either surveying a relatively small number of soils in detail or analyzing a larger number of soils with techniques that provide little detail about the phylogenetic structure of the bacterial communities. Here we used a bar-coded pyrosequencing technique to characterize bacterial communities in 88 soils from across North and South America, obtaining an average of 1,501 sequences per soil. We found that overall bacterial community composition, as measured by pairwise UniFrac distances, was significantly correlated with differences in soil pH (r = 0.79), largely driven by changes in the relative abundances of Acidobacteria, Actinobacteria, and Bacteroidetes across the range of soil pHs. In addition, soil pH explains a significant portion of the variability associated with observed changes in the phylogenetic structure within each dominant lineage. The overall phylogenetic diversity of the bacterial communities was also correlated with soil pH (R2 = 0.50), with peak diversity in soils with near-neutral pHs. Together, these results suggest that the structure of soil bacterial communities is predictable, to some degree, across larger spatial scales, and the effect of soil pH on bacterial community composition is evident at even relatively coarse levels of taxonomic resolution."	ENVO:forest soil	TCAG	Pyrosequencing_based_assessment_of_soil_pH_as_a_predictor_of_soil_bacterial_community_structure_at_the_continental_scale	CL1_V2	0	bacterial biogeography	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19502440	NA	NA	NA	Chris Lauber	CCME	GAZ:United States of America	sandy loam	lauber_88_soils	0-0.05	NA	NA	35	NA	n	NA	NA	Lauber_88_soils	pyrosequencing	NA	NA	-420	soil metagenome	NA	ENVO:Temperate broadleaf and mixed forest biome	FLX	CL1_V2	NA	1250	5.68	NA	Lauber88Soils	Pyrosequencing_based_assessment_of_soil_pH_as_a_predictor_of_soil_bacterial_community_structure_at_the_continental_scale	410658	CCME	ultisol	NA	NA	103	19502440	lauber_88_soils	NA	y	NA	CL1_V2	ENVO:soil	16S rRNA	NA	NA	V2	CCME	V2	NA	NA	Engencore	NA	34.61666667	NA	NA	soil metagenome
CL3.141664	ACAGTGCTTCAT	CATGCTGCCTCCCGTAGGAGT	chris.lauber@gmail.com	NA	NA	ENVO:forest_soil	Soil	source	NA	n	5/28/08	Pyrosequencing-Based Assessment of Soil pH as a Predictor of Soil Bacterial Community Structure at the Continental Scale	NA	FA6P1OK	NA	mimarks-survey	NA	NA	NA	12.1	0	NA	"Calhoun Experimental Forest, SC USA"	150	2008	NA	0	NA	NA	NA	24.3	15.9	NA	NA	NA	CL3	1	"0.5, g"	NA	NA	targeted gene survey	4.145	NA	-81.66666667	y	0	CCME	NA	NA	NA	"Soils harbor enormously diverse bacterial populations, and soil bacterial communities can vary greatly in composition across space. However, our understanding of the specific changes in soil bacterial community structure that occur across larger spatial scales is limited because most previous work has focused on either surveying a relatively small number of soils in detail or analyzing a larger number of soils with techniques that provide little detail about the phylogenetic structure of the bacterial communities. Here we used a bar-coded pyrosequencing technique to characterize bacterial communities in 88 soils from across North and South America, obtaining an average of 1,501 sequences per soil. We found that overall bacterial community composition, as measured by pairwise UniFrac distances, was significantly correlated with differences in soil pH (r = 0.79), largely driven by changes in the relative abundances of Acidobacteria, Actinobacteria, and Bacteroidetes across the range of soil pHs. In addition, soil pH explains a significant portion of the variability associated with observed changes in the phylogenetic structure within each dominant lineage. The overall phylogenetic diversity of the bacterial communities was also correlated with soil pH (R2 = 0.50), with peak diversity in soils with near-neutral pHs. Together, these results suggest that the structure of soil bacterial communities is predictable, to some degree, across larger spatial scales, and the effect of soil pH on bacterial community composition is evident at even relatively coarse levels of taxonomic resolution."	ENVO:forest soil	TCAG	Pyrosequencing_based_assessment_of_soil_pH_as_a_predictor_of_soil_bacterial_community_structure_at_the_continental_scale	CL3_V2	0	bacterial biogeography	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19502440	NA	NA	NA	Chris Lauber	CCME	GAZ:United States of America	loamy sand	lauber_88_soils	0-0.05	NA	NA	19	NA	n	NA	NA	Lauber_88_soils	pyrosequencing	NA	NA	-420	soil metagenome	NA	ENVO:Temperate broadleaf and mixed forest biome	FLX	CL3_V2	NA	1250	4.89	NA	Lauber88Soils	Pyrosequencing_based_assessment_of_soil_pH_as_a_predictor_of_soil_bacterial_community_structure_at_the_continental_scale	410658	CCME	ultisol	NA	NA	103	19502440	lauber_88_soils	NA	y	NA	CL3_V2	ENVO:soil	16S rRNA	NA	NA	V2	CCME	V2	NA	NA	Engencore	NA	34.61666667	NA	NA	soil metagenome
CO1.141714	ACATGATCGTTC	CATGCTGCCTCCCGTAGGAGT	chris.lauber@gmail.com	NA	NA	ENVO:forest_soil	Soil	source	NA	n	5/28/08	Pyrosequencing-Based Assessment of Soil pH as a Predictor of Soil Bacterial Community Structure at the Continental Scale	NA	FA6P1OK	NA	mimarks-survey	NA	NA	NA	15.9	0	NA	"Fort Collins, CO USA"	3800	2008	NA	0	NA	NA	NA	19.1	-3	NA	NA	NA	CO1	1	"0.5, g"	NA	NA	targeted gene survey	4.477	NA	-105.7	y	0	CCME	NA	NA	NA	"Soils harbor enormously diverse bacterial populations, and soil bacterial communities can vary greatly in composition across space. However, our understanding of the specific changes in soil bacterial community structure that occur across larger spatial scales is limited because most previous work has focused on either surveying a relatively small number of soils in detail or analyzing a larger number of soils with techniques that provide little detail about the phylogenetic structure of the bacterial communities. Here we used a bar-coded pyrosequencing technique to characterize bacterial communities in 88 soils from across North and South America, obtaining an average of 1,501 sequences per soil. We found that overall bacterial community composition, as measured by pairwise UniFrac distances, was significantly correlated with differences in soil pH (r = 0.79), largely driven by changes in the relative abundances of Acidobacteria, Actinobacteria, and Bacteroidetes across the range of soil pHs. In addition, soil pH explains a significant portion of the variability associated with observed changes in the phylogenetic structure within each dominant lineage. The overall phylogenetic diversity of the bacterial communities was also correlated with soil pH (R2 = 0.50), with peak diversity in soils with near-neutral pHs. Together, these results suggest that the structure of soil bacterial communities is predictable, to some degree, across larger spatial scales, and the effect of soil pH on bacterial community composition is evident at even relatively coarse levels of taxonomic resolution."	ENVO:forest soil	TCAG	Pyrosequencing_based_assessment_of_soil_pH_as_a_predictor_of_soil_bacterial_community_structure_at_the_continental_scale	CO1_V2	0	bacterial biogeography	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19502440	NA	NA	NA	Chris Lauber	CCME	GAZ:United States of America	sand	lauber_88_soils	0-0.05	NA	NA	14	NA	n	NA	NA	Lauber_88_soils	pyrosequencing	NA	NA	-309	soil metagenome	NA	ENVO:forest	FLX	CO1_V2	NA	600	6.13	NA	Lauber88Soils	Pyrosequencing_based_assessment_of_soil_pH_as_a_predictor_of_soil_bacterial_community_structure_at_the_continental_scale	410658	CCME	inceptisol	NA	NA	103	19502440	lauber_88_soils	NA	y	NA	CO1_V2	ENVO:soil	16S rRNA	NA	NA	V2	CCME	V2	NA	NA	Engencore	NA	40.4	NA	NA	soil metagenome
CO2.141657	ACATGTCACGTG	CATGCTGCCTCCCGTAGGAGT	chris.lauber@gmail.com	NA	NA	ENVO:forest_soil	Soil	source	NA	n	5/28/08	Pyrosequencing-Based Assessment of Soil pH as a Predictor of Soil Bacterial Community Structure at the Continental Scale	NA	FA6P1OK	NA	mimarks-survey	NA	NA	NA	18.1	0	NA	"Fort Collins, CO USA"	2400	2008	NA	0	NA	NA	NA	31.8	6.1	NA	NA	NA	CO2	1	"0.5, g"	NA	NA	targeted gene survey	9.223	NA	-105.3333333	y	0	CCME	NA	NA	NA	"Soils harbor enormously diverse bacterial populations, and soil bacterial communities can vary greatly in composition across space. However, our understanding of the specific changes in soil bacterial community structure that occur across larger spatial scales is limited because most previous work has focused on either surveying a relatively small number of soils in detail or analyzing a larger number of soils with techniques that provide little detail about the phylogenetic structure of the bacterial communities. Here we used a bar-coded pyrosequencing technique to characterize bacterial communities in 88 soils from across North and South America, obtaining an average of 1,501 sequences per soil. We found that overall bacterial community composition, as measured by pairwise UniFrac distances, was significantly correlated with differences in soil pH (r = 0.79), largely driven by changes in the relative abundances of Acidobacteria, Actinobacteria, and Bacteroidetes across the range of soil pHs. In addition, soil pH explains a significant portion of the variability associated with observed changes in the phylogenetic structure within each dominant lineage. The overall phylogenetic diversity of the bacterial communities was also correlated with soil pH (R2 = 0.50), with peak diversity in soils with near-neutral pHs. Together, these results suggest that the structure of soil bacterial communities is predictable, to some degree, across larger spatial scales, and the effect of soil pH on bacterial community composition is evident at even relatively coarse levels of taxonomic resolution."	ENVO:forest soil	TCAG	Pyrosequencing_based_assessment_of_soil_pH_as_a_predictor_of_soil_bacterial_community_structure_at_the_continental_scale	CO2_V2	0	bacterial biogeography	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19502440	NA	NA	NA	Chris Lauber	CCME	GAZ:United States of America	sandy loam	lauber_88_soils	0-0.05	NA	NA	24	NA	n	NA	NA	Lauber_88_soils	pyrosequencing	NA	NA	104	soil metagenome	NA	ENVO:forest	FLX	CO2_V2	NA	350	5.68	NA	Lauber88Soils	Pyrosequencing_based_assessment_of_soil_pH_as_a_predictor_of_soil_bacterial_community_structure_at_the_continental_scale	410658	CCME	alfisol	NA	NA	103	19502440	lauber_88_soils	NA	y	NA	CO2_V2	ENVO:soil	16S rRNA	NA	NA	V2	CCME	V2	NA	NA	Engencore	NA	40.58333333	NA	NA	soil metagenome
DF1.141684	ACCAGACGATGC	CATGCTGCCTCCCGTAGGAGT	chris.lauber@gmail.com	NA	NA	ENVO:forest_soil	Soil	source	NA	n	5/28/08	Pyrosequencing-Based Assessment of Soil pH as a Predictor of Soil Bacterial Community Structure at the Continental Scale	NA	FA6P1OK	NA	mimarks-survey	NA	NA	NA	27.8	0	NA	"Duke Forest, NC USA"	163	2008	NA	0	NA	NA	NA	37.7	14.6	NA	NA	NA	DF1	1	"0.5, g"	NA	NA	targeted gene survey	13.201	NA	-79.08333333	y	0	CCME	NA	NA	NA	"Soils harbor enormously diverse bacterial populations, and soil bacterial communities can vary greatly in composition across space. However, our understanding of the specific changes in soil bacterial community structure that occur across larger spatial scales is limited because most previous work has focused on either surveying a relatively small number of soils in detail or analyzing a larger number of soils with techniques that provide little detail about the phylogenetic structure of the bacterial communities. Here we used a bar-coded pyrosequencing technique to characterize bacterial communities in 88 soils from across North and South America, obtaining an average of 1,501 sequences per soil. We found that overall bacterial community composition, as measured by pairwise UniFrac distances, was significantly correlated with differences in soil pH (r = 0.79), largely driven by changes in the relative abundances of Acidobacteria, Actinobacteria, and Bacteroidetes across the range of soil pHs. In addition, soil pH explains a significant portion of the variability associated with observed changes in the phylogenetic structure within each dominant lineage. The overall phylogenetic diversity of the bacterial communities was also correlated with soil pH (R2 = 0.50), with peak diversity in soils with near-neutral pHs. Together, these results suggest that the structure of soil bacterial communities is predictable, to some degree, across larger spatial scales, and the effect of soil pH on bacterial community composition is evident at even relatively coarse levels of taxonomic resolution."	ENVO:forest soil	TCAG	Pyrosequencing_based_assessment_of_soil_pH_as_a_predictor_of_soil_bacterial_community_structure_at_the_continental_scale	DF1_V2	0	bacterial biogeography	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19502440	NA	NA	NA	Chris Lauber	CCME	GAZ:United States of America	sandy loam	lauber_88_soils	0-0.05	NA	NA	43	NA	n	NA	NA	Lauber_88_soils	pyrosequencing	NA	NA	-328	soil metagenome	NA	ENVO:Temperate broadleaf and mixed forest biome	FLX	DF1_V2	NA	1100	5.37	NA	Lauber88Soils	Pyrosequencing_based_assessment_of_soil_pH_as_a_predictor_of_soil_bacterial_community_structure_at_the_continental_scale	410658	CCME	alfisol	NA	NA	103	19502440	lauber_88_soils	NA	y	NA	DF1_V2	ENVO:soil	16S rRNA	NA	NA	V2	CCME	V2	NA	NA	Engencore	NA	35.96666667	NA	NA	soil metagenome
DF2.141726	ACCAGCGACTAG	CATGCTGCCTCCCGTAGGAGT	chris.lauber@gmail.com	NA	NA	ENVO:forest_soil	Soil	source	NA	n	5/28/08	Pyrosequencing-Based Assessment of Soil pH as a Predictor of Soil Bacterial Community Structure at the Continental Scale	NA	FA6P1OK	NA	mimarks-survey	NA	NA	NA	54.5	0	NA	"Duke Forest, NC USA"	163	2008	NA	0	NA	NA	NA	18.3	14.6	NA	NA	NA	DF2	1	"0.5, g"	NA	NA	targeted gene survey	20.411	NA	-79.08333333	y	0	CCME	NA	NA	NA	"Soils harbor enormously diverse bacterial populations, and soil bacterial communities can vary greatly in composition across space. However, our understanding of the specific changes in soil bacterial community structure that occur across larger spatial scales is limited because most previous work has focused on either surveying a relatively small number of soils in detail or analyzing a larger number of soils with techniques that provide little detail about the phylogenetic structure of the bacterial communities. Here we used a bar-coded pyrosequencing technique to characterize bacterial communities in 88 soils from across North and South America, obtaining an average of 1,501 sequences per soil. We found that overall bacterial community composition, as measured by pairwise UniFrac distances, was significantly correlated with differences in soil pH (r = 0.79), largely driven by changes in the relative abundances of Acidobacteria, Actinobacteria, and Bacteroidetes across the range of soil pHs. In addition, soil pH explains a significant portion of the variability associated with observed changes in the phylogenetic structure within each dominant lineage. The overall phylogenetic diversity of the bacterial communities was also correlated with soil pH (R2 = 0.50), with peak diversity in soils with near-neutral pHs. Together, these results suggest that the structure of soil bacterial communities is predictable, to some degree, across larger spatial scales, and the effect of soil pH on bacterial community composition is evident at even relatively coarse levels of taxonomic resolution."	ENVO:forest soil	TCAG	Pyrosequencing_based_assessment_of_soil_pH_as_a_predictor_of_soil_bacterial_community_structure_at_the_continental_scale	DF2_V2	0	bacterial biogeography	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19502440	NA	NA	NA	Chris Lauber	CCME	GAZ:United States of America	loam	lauber_88_soils	0-0.05	NA	NA	53	NA	n	NA	NA	Lauber_88_soils	pyrosequencing	NA	NA	-328	soil metagenome	NA	ENVO:Temperate broadleaf and mixed forest biome	FLX	DF2_V2	NA	1100	6.84	NA	Lauber88Soils	Pyrosequencing_based_assessment_of_soil_pH_as_a_predictor_of_soil_bacterial_community_structure_at_the_continental_scale	410658	CCME	alfisol	NA	NA	103	19502440	lauber_88_soils	NA	y	NA	DF2_V2	ENVO:soil	16S rRNA	NA	NA	V2	CCME	V2	NA	NA	Engencore	NA	35.96666667	NA	NA	soil metagenome
DF3.141696	ACCGCAGAGTCA	CATGCTGCCTCCCGTAGGAGT	chris.lauber@gmail.com	NA	NA	ENVO:forest_soil	Soil	source	NA	n	5/28/08	Pyrosequencing-Based Assessment of Soil pH as a Predictor of Soil Bacterial Community Structure at the Continental Scale	NA	FA6P1OK	NA	mimarks-survey	NA	NA	NA	17	0	NA	"Duke Forest, NC USA"	150	2008	NA	0	NA	NA	NA	25.8	14.6	NA	NA	NA	DF3	1	"0.5, g"	NA	NA	targeted gene survey	4.149	NA	-79.08333333	y	0	CCME	NA	NA	NA	"Soils harbor enormously diverse bacterial populations, and soil bacterial communities can vary greatly in composition across space. However, our understanding of the specific changes in soil bacterial community structure that occur across larger spatial scales is limited because most previous work has focused on either surveying a relatively small number of soils in detail or analyzing a larger number of soils with techniques that provide little detail about the phylogenetic structure of the bacterial communities. Here we used a bar-coded pyrosequencing technique to characterize bacterial communities in 88 soils from across North and South America, obtaining an average of 1,501 sequences per soil. We found that overall bacterial community composition, as measured by pairwise UniFrac distances, was significantly correlated with differences in soil pH (r = 0.79), largely driven by changes in the relative abundances of Acidobacteria, Actinobacteria, and Bacteroidetes across the range of soil pHs. In addition, soil pH explains a significant portion of the variability associated with observed changes in the phylogenetic structure within each dominant lineage. The overall phylogenetic diversity of the bacterial communities was also correlated with soil pH (R2 = 0.50), with peak diversity in soils with near-neutral pHs. Together, these results suggest that the structure of soil bacterial communities is predictable, to some degree, across larger spatial scales, and the effect of soil pH on bacterial community composition is evident at even relatively coarse levels of taxonomic resolution."	ENVO:forest soil	TCAG	Pyrosequencing_based_assessment_of_soil_pH_as_a_predictor_of_soil_bacterial_community_structure_at_the_continental_scale	DF3_V2	0	bacterial biogeography	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19502440	NA	NA	NA	Chris Lauber	CCME	GAZ:United States of America	loamy sand	lauber_88_soils	0-0.05	NA	NA	20	NA	n	NA	NA	Lauber_88_soils	pyrosequencing	NA	NA	-328	soil metagenome	NA	ENVO:Temperate broadleaf and mixed forest biome	FLX	DF3_V2	NA	1100	5.05	NA	Lauber88Soils	Pyrosequencing_based_assessment_of_soil_pH_as_a_predictor_of_soil_bacterial_community_structure_at_the_continental_scale	410658	CCME	alfisol	NA	NA	103	19502440	lauber_88_soils	NA	y	NA	DF3_V2	ENVO:soil	16S rRNA	NA	NA	V2	CCME	V2	NA	NA	Engencore	NA	35.96666667	NA	NA	soil metagenome
GB2.141732	ACCTGTCTCTCT	CATGCTGCCTCCCGTAGGAGT	chris.lauber@gmail.com	NA	NA	ENVO:forest_soil	Soil	source	NA	n	5/28/08	Pyrosequencing-Based Assessment of Soil pH as a Predictor of Soil Bacterial Community Structure at the Continental Scale	NA	FA6P1OK	NA	mimarks-survey	NA	NA	NA	68.9	0	NA	"Great Basin Experimental Range, UT USA"	3290	2008	NA	0	NA	NA	NA	13.1	2	NA	NA	NA	GB2	1	"0.5, g"	NA	NA	targeted gene survey	5.375	NA	-111.4666667	y	0	CCME	NA	NA	NA	"Soils harbor enormously diverse bacterial populations, and soil bacterial communities can vary greatly in composition across space. However, our understanding of the specific changes in soil bacterial community structure that occur across larger spatial scales is limited because most previous work has focused on either surveying a relatively small number of soils in detail or analyzing a larger number of soils with techniques that provide little detail about the phylogenetic structure of the bacterial communities. Here we used a bar-coded pyrosequencing technique to characterize bacterial communities in 88 soils from across North and South America, obtaining an average of 1,501 sequences per soil. We found that overall bacterial community composition, as measured by pairwise UniFrac distances, was significantly correlated with differences in soil pH (r = 0.79), largely driven by changes in the relative abundances of Acidobacteria, Actinobacteria, and Bacteroidetes across the range of soil pHs. In addition, soil pH explains a significant portion of the variability associated with observed changes in the phylogenetic structure within each dominant lineage. The overall phylogenetic diversity of the bacterial communities was also correlated with soil pH (R2 = 0.50), with peak diversity in soils with near-neutral pHs. Together, these results suggest that the structure of soil bacterial communities is predictable, to some degree, across larger spatial scales, and the effect of soil pH on bacterial community composition is evident at even relatively coarse levels of taxonomic resolution."	ENVO:forest soil	TCAG	Pyrosequencing_based_assessment_of_soil_pH_as_a_predictor_of_soil_bacterial_community_structure_at_the_continental_scale	GB2_V2	0	bacterial biogeography	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19502440	NA	NA	NA	Chris Lauber	CCME	GAZ:United States of America	loam	lauber_88_soils	0-0.05	NA	NA	65	NA	n	NA	NA	Lauber_88_soils	pyrosequencing	NA	NA	-112	soil metagenome	NA	ENVO:forest	FLX	GB2_V2	NA	400	7.57	NA	Lauber88Soils	Pyrosequencing_based_assessment_of_soil_pH_as_a_predictor_of_soil_bacterial_community_structure_at_the_continental_scale	410658	CCME	mollisol	NA	NA	103	19502440	lauber_88_soils	NA	y	NA	GB2_V2	ENVO:soil	16S rRNA	NA	NA	V2	CCME	V2	NA	NA	Engencore	NA	39.31666667	NA	NA	soil metagenome
GB3.141652	ACGACGTCTTAG	CATGCTGCCTCCCGTAGGAGT	chris.lauber@gmail.com	NA	NA	ENVO:forest_soil	Soil	source	NA	n	5/28/08	Pyrosequencing-Based Assessment of Soil pH as a Predictor of Soil Bacterial Community Structure at the Continental Scale	NA	FA6P1OK	NA	mimarks-survey	NA	NA	NA	57.1	0	NA	"Great Basin Experimental Range, UT USA"	3270	2008	NA	0	NA	NA	NA	14.3	2	NA	NA	NA	GB3	1	"0.5, g"	NA	NA	targeted gene survey	6.244	NA	-111.4833333	y	0	CCME	NA	NA	NA	"Soils harbor enormously diverse bacterial populations, and soil bacterial communities can vary greatly in composition across space. However, our understanding of the specific changes in soil bacterial community structure that occur across larger spatial scales is limited because most previous work has focused on either surveying a relatively small number of soils in detail or analyzing a larger number of soils with techniques that provide little detail about the phylogenetic structure of the bacterial communities. Here we used a bar-coded pyrosequencing technique to characterize bacterial communities in 88 soils from across North and South America, obtaining an average of 1,501 sequences per soil. We found that overall bacterial community composition, as measured by pairwise UniFrac distances, was significantly correlated with differences in soil pH (r = 0.79), largely driven by changes in the relative abundances of Acidobacteria, Actinobacteria, and Bacteroidetes across the range of soil pHs. In addition, soil pH explains a significant portion of the variability associated with observed changes in the phylogenetic structure within each dominant lineage. The overall phylogenetic diversity of the bacterial communities was also correlated with soil pH (R2 = 0.50), with peak diversity in soils with near-neutral pHs. Together, these results suggest that the structure of soil bacterial communities is predictable, to some degree, across larger spatial scales, and the effect of soil pH on bacterial community composition is evident at even relatively coarse levels of taxonomic resolution."	ENVO:forest soil	TCAG	Pyrosequencing_based_assessment_of_soil_pH_as_a_predictor_of_soil_bacterial_community_structure_at_the_continental_scale	GB3_V2	0	bacterial biogeography	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19502440	NA	NA	NA	Chris Lauber	CCME	GAZ:United States of America	clay loam	lauber_88_soils	0-0.05	NA	NA	76	NA	n	NA	NA	Lauber_88_soils	pyrosequencing	NA	NA	-112	soil metagenome	NA	ENVO:forest	FLX	GB3_V2	NA	400	7.18	NA	Lauber88Soils	Pyrosequencing_based_assessment_of_soil_pH_as_a_predictor_of_soil_bacterial_community_structure_at_the_continental_scale	410658	CCME	mollisol	NA	NA	103	19502440	lauber_88_soils	NA	y	NA	GB3_V2	ENVO:soil	16S rRNA	NA	NA	V2	CCME	V2	NA	NA	Engencore	NA	39.31666667	NA	NA	soil metagenome
HF1.141663	ACGATGCGACCA	CATGCTGCCTCCCGTAGGAGT	chris.lauber@gmail.com	NA	NA	ENVO:forest_soil	Soil	source	NA	n	5/28/08	Pyrosequencing-Based Assessment of Soil pH as a Predictor of Soil Bacterial Community Structure at the Continental Scale	NA	FA6P1OK	NA	mimarks-survey	NA	NA	NA	127.6	0	NA	"Harvard Forest, MA USA"	300	2008	NA	0	NA	NA	NA	22.5	7	NA	NA	NA	HF1	1	"0.5, g"	NA	NA	targeted gene survey	14.412	NA	-72.16666667	y	0	CCME	NA	NA	NA	"Soils harbor enormously diverse bacterial populations, and soil bacterial communities can vary greatly in composition across space. However, our understanding of the specific changes in soil bacterial community structure that occur across larger spatial scales is limited because most previous work has focused on either surveying a relatively small number of soils in detail or analyzing a larger number of soils with techniques that provide little detail about the phylogenetic structure of the bacterial communities. Here we used a bar-coded pyrosequencing technique to characterize bacterial communities in 88 soils from across North and South America, obtaining an average of 1,501 sequences per soil. We found that overall bacterial community composition, as measured by pairwise UniFrac distances, was significantly correlated with differences in soil pH (r = 0.79), largely driven by changes in the relative abundances of Acidobacteria, Actinobacteria, and Bacteroidetes across the range of soil pHs. In addition, soil pH explains a significant portion of the variability associated with observed changes in the phylogenetic structure within each dominant lineage. The overall phylogenetic diversity of the bacterial communities was also correlated with soil pH (R2 = 0.50), with peak diversity in soils with near-neutral pHs. Together, these results suggest that the structure of soil bacterial communities is predictable, to some degree, across larger spatial scales, and the effect of soil pH on bacterial community composition is evident at even relatively coarse levels of taxonomic resolution."	ENVO:forest soil	TCAG	Pyrosequencing_based_assessment_of_soil_pH_as_a_predictor_of_soil_bacterial_community_structure_at_the_continental_scale	HF1_V2	0	bacterial biogeography	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19502440	NA	NA	NA	Chris Lauber	CCME	GAZ:United States of America	sandy loam	lauber_88_soils	0-0.05	NA	NA	33	NA	n	NA	NA	Lauber_88_soils	pyrosequencing	NA	NA	-484	soil metagenome	NA	ENVO:Temperate broadleaf and mixed forest biome	FLX	HF1_V2	NA	1100	4.25	NA	Lauber88Soils	Pyrosequencing_based_assessment_of_soil_pH_as_a_predictor_of_soil_bacterial_community_structure_at_the_continental_scale	410658	CCME	inceptisol	NA	NA	103	19502440	lauber_88_soils	NA	y	NA	HF1_V2	ENVO:soil	16S rRNA	NA	NA	V2	CCME	V2	NA	NA	Engencore	NA	42.5	NA	NA	soil metagenome
HF2.141686	ACGCAACTGCTA	CATGCTGCCTCCCGTAGGAGT	chris.lauber@gmail.com	NA	NA	ENVO:forest_soil	Soil	source	NA	n	5/28/08	Pyrosequencing-Based Assessment of Soil pH as a Predictor of Soil Bacterial Community Structure at the Continental Scale	NA	FA6P1OK	NA	mimarks-survey	NA	NA	NA	95.5	0	NA	"Harvard Forest, MA USA"	300	2008	NA	0	NA	NA	NA	20.9	7	NA	NA	NA	HF2	1	"0.5, g"	NA	NA	targeted gene survey	10.001	NA	-72.16666667	y	0	CCME	NA	NA	NA	"Soils harbor enormously diverse bacterial populations, and soil bacterial communities can vary greatly in composition across space. However, our understanding of the specific changes in soil bacterial community structure that occur across larger spatial scales is limited because most previous work has focused on either surveying a relatively small number of soils in detail or analyzing a larger number of soils with techniques that provide little detail about the phylogenetic structure of the bacterial communities. Here we used a bar-coded pyrosequencing technique to characterize bacterial communities in 88 soils from across North and South America, obtaining an average of 1,501 sequences per soil. We found that overall bacterial community composition, as measured by pairwise UniFrac distances, was significantly correlated with differences in soil pH (r = 0.79), largely driven by changes in the relative abundances of Acidobacteria, Actinobacteria, and Bacteroidetes across the range of soil pHs. In addition, soil pH explains a significant portion of the variability associated with observed changes in the phylogenetic structure within each dominant lineage. The overall phylogenetic diversity of the bacterial communities was also correlated with soil pH (R2 = 0.50), with peak diversity in soils with near-neutral pHs. Together, these results suggest that the structure of soil bacterial communities is predictable, to some degree, across larger spatial scales, and the effect of soil pH on bacterial community composition is evident at even relatively coarse levels of taxonomic resolution."	ENVO:forest soil	TCAG	Pyrosequencing_based_assessment_of_soil_pH_as_a_predictor_of_soil_bacterial_community_structure_at_the_continental_scale	HF2_V2	0	bacterial biogeography	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19502440	NA	NA	NA	Chris Lauber	CCME	GAZ:United States of America	sandy loam	lauber_88_soils	0-0.05	NA	NA	38	NA	n	NA	NA	Lauber_88_soils	pyrosequencing	NA	NA	-484	soil metagenome	NA	ENVO:Temperate broadleaf and mixed forest biome	FLX	HF2_V2	NA	1100	3.98	NA	Lauber88Soils	Pyrosequencing_based_assessment_of_soil_pH_as_a_predictor_of_soil_bacterial_community_structure_at_the_continental_scale	410658	CCME	inceptisol	NA	NA	103	19502440	lauber_88_soils	NA	y	NA	HF2_V2	ENVO:soil	16S rRNA	NA	NA	V2	CCME	V2	NA	NA	Engencore	NA	42.5	NA	NA	soil metagenome
HJ1.141718	ACGGATCGTCAG	CATGCTGCCTCCCGTAGGAGT	chris.lauber@gmail.com	NA	NA	ENVO:forest_soil	Soil	source	NA	n	5/28/08	Pyrosequencing-Based Assessment of Soil pH as a Predictor of Soil Bacterial Community Structure at the Continental Scale	NA	FA6P1OK	NA	mimarks-survey	NA	NA	NA	69.5	0	NA	"H.J. Andrews Experimental Forest, OR USA"	700	2008	NA	0	NA	NA	NA	36.6	9.4	NA	NA	NA	HJ1	1	"0.5, g"	NA	NA	targeted gene survey	5.079	NA	-122.15	y	0	CCME	NA	NA	NA	"Soils harbor enormously diverse bacterial populations, and soil bacterial communities can vary greatly in composition across space. However, our understanding of the specific changes in soil bacterial community structure that occur across larger spatial scales is limited because most previous work has focused on either surveying a relatively small number of soils in detail or analyzing a larger number of soils with techniques that provide little detail about the phylogenetic structure of the bacterial communities. Here we used a bar-coded pyrosequencing technique to characterize bacterial communities in 88 soils from across North and South America, obtaining an average of 1,501 sequences per soil. We found that overall bacterial community composition, as measured by pairwise UniFrac distances, was significantly correlated with differences in soil pH (r = 0.79), largely driven by changes in the relative abundances of Acidobacteria, Actinobacteria, and Bacteroidetes across the range of soil pHs. In addition, soil pH explains a significant portion of the variability associated with observed changes in the phylogenetic structure within each dominant lineage. The overall phylogenetic diversity of the bacterial communities was also correlated with soil pH (R2 = 0.50), with peak diversity in soils with near-neutral pHs. Together, these results suggest that the structure of soil bacterial communities is predictable, to some degree, across larger spatial scales, and the effect of soil pH on bacterial community composition is evident at even relatively coarse levels of taxonomic resolution."	ENVO:forest soil	TCAG	Pyrosequencing_based_assessment_of_soil_pH_as_a_predictor_of_soil_bacterial_community_structure_at_the_continental_scale	HJ1_V2	0	bacterial biogeography	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19502440	NA	NA	NA	Chris Lauber	CCME	GAZ:United States of America	sandy loam	lauber_88_soils	0-0.05	NA	NA	41	NA	n	NA	NA	Lauber_88_soils	pyrosequencing	NA	NA	-1507	soil metagenome	NA	ENVO:Temperate broadleaf and mixed forest biome	FLX	HJ1_V2	NA	2000	5.41	NA	Lauber88Soils	Pyrosequencing_based_assessment_of_soil_pH_as_a_predictor_of_soil_bacterial_community_structure_at_the_continental_scale	410658	CCME	andisol	NA	NA	103	19502440	lauber_88_soils	NA	y	NA	HJ1_V2	ENVO:soil	16S rRNA	NA	NA	V2	CCME	V2	NA	NA	Engencore	NA	44.21666667	NA	NA	soil metagenome
HJ2.141717	ACGGTGAGTGTC	CATGCTGCCTCCCGTAGGAGT	chris.lauber@gmail.com	NA	NA	ENVO:forest_soil	Soil	source	NA	n	5/28/08	Pyrosequencing-Based Assessment of Soil pH as a Predictor of Soil Bacterial Community Structure at the Continental Scale	NA	FA6P1OK	NA	mimarks-survey	NA	NA	NA	76.1	0	NA	"H.J. Andrews Experimental Forest, OR USA"	700	2008	NA	0	NA	NA	NA	26.1	9.4	NA	NA	NA	HJ2	1	"0.5, g"	NA	NA	targeted gene survey	4.532	NA	-122.15	y	0	CCME	NA	NA	NA	"Soils harbor enormously diverse bacterial populations, and soil bacterial communities can vary greatly in composition across space. However, our understanding of the specific changes in soil bacterial community structure that occur across larger spatial scales is limited because most previous work has focused on either surveying a relatively small number of soils in detail or analyzing a larger number of soils with techniques that provide little detail about the phylogenetic structure of the bacterial communities. Here we used a bar-coded pyrosequencing technique to characterize bacterial communities in 88 soils from across North and South America, obtaining an average of 1,501 sequences per soil. We found that overall bacterial community composition, as measured by pairwise UniFrac distances, was significantly correlated with differences in soil pH (r = 0.79), largely driven by changes in the relative abundances of Acidobacteria, Actinobacteria, and Bacteroidetes across the range of soil pHs. In addition, soil pH explains a significant portion of the variability associated with observed changes in the phylogenetic structure within each dominant lineage. The overall phylogenetic diversity of the bacterial communities was also correlated with soil pH (R2 = 0.50), with peak diversity in soils with near-neutral pHs. Together, these results suggest that the structure of soil bacterial communities is predictable, to some degree, across larger spatial scales, and the effect of soil pH on bacterial community composition is evident at even relatively coarse levels of taxonomic resolution."	ENVO:forest soil	TCAG	Pyrosequencing_based_assessment_of_soil_pH_as_a_predictor_of_soil_bacterial_community_structure_at_the_continental_scale	HJ2_V2	0	bacterial biogeography	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19502440	NA	NA	NA	Chris Lauber	CCME	GAZ:United States of America	sandy loam	lauber_88_soils	0-0.05	NA	NA	47	NA	n	NA	NA	Lauber_88_soils	pyrosequencing	NA	NA	-1507	soil metagenome	NA	ENVO:Temperate broadleaf and mixed forest biome	FLX	HJ2_V2	NA	2000	5.36	NA	Lauber88Soils	Pyrosequencing_based_assessment_of_soil_pH_as_a_predictor_of_soil_bacterial_community_structure_at_the_continental_scale	410658	CCME	andisol	NA	NA	103	19502440	lauber_88_soils	NA	y	NA	HJ2_V2	ENVO:soil	16S rRNA	NA	NA	V2	CCME	V2	NA	NA	Engencore	NA	44.21666667	NA	NA	soil metagenome
IT1.141711	ACGTGAGAGAAT	CATGCTGCCTCCCGTAGGAGT	chris.lauber@gmail.com	NA	NA	ENVO:forest_soil	Soil	source	NA	n	5/28/08	Pyrosequencing-Based Assessment of Soil pH as a Predictor of Soil Bacterial Community Structure at the Continental Scale	NA	FA6P1OK	NA	mimarks-survey	NA	NA	NA	63.1	0	NA	"Itasca Lake State Park, MN USA"	550	2008	NA	0	NA	NA	NA	22.5	3	NA	NA	NA	IT1	1	"0.5, g"	NA	NA	targeted gene survey	16.079	NA	-95.16666667	y	0	CCME	NA	NA	NA	"Soils harbor enormously diverse bacterial populations, and soil bacterial communities can vary greatly in composition across space. However, our understanding of the specific changes in soil bacterial community structure that occur across larger spatial scales is limited because most previous work has focused on either surveying a relatively small number of soils in detail or analyzing a larger number of soils with techniques that provide little detail about the phylogenetic structure of the bacterial communities. Here we used a bar-coded pyrosequencing technique to characterize bacterial communities in 88 soils from across North and South America, obtaining an average of 1,501 sequences per soil. We found that overall bacterial community composition, as measured by pairwise UniFrac distances, was significantly correlated with differences in soil pH (r = 0.79), largely driven by changes in the relative abundances of Acidobacteria, Actinobacteria, and Bacteroidetes across the range of soil pHs. In addition, soil pH explains a significant portion of the variability associated with observed changes in the phylogenetic structure within each dominant lineage. The overall phylogenetic diversity of the bacterial communities was also correlated with soil pH (R2 = 0.50), with peak diversity in soils with near-neutral pHs. Together, these results suggest that the structure of soil bacterial communities is predictable, to some degree, across larger spatial scales, and the effect of soil pH on bacterial community composition is evident at even relatively coarse levels of taxonomic resolution."	ENVO:forest soil	TCAG	Pyrosequencing_based_assessment_of_soil_pH_as_a_predictor_of_soil_bacterial_community_structure_at_the_continental_scale	IT1_V2	0	bacterial biogeography	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19502440	NA	NA	NA	Chris Lauber	CCME	GAZ:United States of America	sandy loam	lauber_88_soils	0-0.05	NA	NA	27	NA	n	NA	NA	Lauber_88_soils	pyrosequencing	NA	NA	-262	soil metagenome	NA	ENVO:Temperate broadleaf and mixed forest biome	FLX	IT1_V2	NA	750	5.78	NA	Lauber88Soils	Pyrosequencing_based_assessment_of_soil_pH_as_a_predictor_of_soil_bacterial_community_structure_at_the_continental_scale	410658	CCME	spodosol	NA	NA	103	19502440	lauber_88_soils	NA	y	NA	IT1_V2	ENVO:soil	16S rRNA	NA	NA	V2	CCME	V2	NA	NA	Engencore	NA	47.16666667	NA	NA	soil metagenome
IT2.141720	ACGTGCCGTAGA	CATGCTGCCTCCCGTAGGAGT	chris.lauber@gmail.com	NA	NA	ENVO:forest_soil	Soil	source	NA	n	5/28/08	Pyrosequencing-Based Assessment of Soil pH as a Predictor of Soil Bacterial Community Structure at the Continental Scale	NA	FA6P1OK	NA	mimarks-survey	NA	NA	NA	39.1	0	NA	"Itasca Lake State Park, MN USA"	550	2008	NA	0	NA	NA	NA	23	3	NA	NA	NA	IT2	1	"0.5, g"	NA	NA	targeted gene survey	16.003	NA	-95.16666667	y	0	CCME	NA	NA	NA	"Soils harbor enormously diverse bacterial populations, and soil bacterial communities can vary greatly in composition across space. However, our understanding of the specific changes in soil bacterial community structure that occur across larger spatial scales is limited because most previous work has focused on either surveying a relatively small number of soils in detail or analyzing a larger number of soils with techniques that provide little detail about the phylogenetic structure of the bacterial communities. Here we used a bar-coded pyrosequencing technique to characterize bacterial communities in 88 soils from across North and South America, obtaining an average of 1,501 sequences per soil. We found that overall bacterial community composition, as measured by pairwise UniFrac distances, was significantly correlated with differences in soil pH (r = 0.79), largely driven by changes in the relative abundances of Acidobacteria, Actinobacteria, and Bacteroidetes across the range of soil pHs. In addition, soil pH explains a significant portion of the variability associated with observed changes in the phylogenetic structure within each dominant lineage. The overall phylogenetic diversity of the bacterial communities was also correlated with soil pH (R2 = 0.50), with peak diversity in soils with near-neutral pHs. Together, these results suggest that the structure of soil bacterial communities is predictable, to some degree, across larger spatial scales, and the effect of soil pH on bacterial community composition is evident at even relatively coarse levels of taxonomic resolution."	ENVO:forest soil	TCAG	Pyrosequencing_based_assessment_of_soil_pH_as_a_predictor_of_soil_bacterial_community_structure_at_the_continental_scale	IT2_V2	0	bacterial biogeography	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19502440	NA	NA	NA	Chris Lauber	CCME	GAZ:United States of America	loamy sand	lauber_88_soils	0-0.05	NA	NA	19	NA	n	NA	NA	Lauber_88_soils	pyrosequencing	NA	NA	-262	soil metagenome	NA	ENVO:Temperate broadleaf and mixed forest biome	FLX	IT2_V2	NA	750	5.42	NA	Lauber88Soils	Pyrosequencing_based_assessment_of_soil_pH_as_a_predictor_of_soil_bacterial_community_structure_at_the_continental_scale	410658	CCME	spodosol	NA	NA	103	19502440	lauber_88_soils	NA	y	NA	IT2_V2	ENVO:soil	16S rRNA	NA	NA	V2	CCME	V2	NA	NA	Engencore	NA	47.18333333	NA	NA	soil metagenome
MP2.141695	ACTGTACGCGTA	CATGCTGCCTCCCGTAGGAGT	chris.lauber@gmail.com	NA	NA	ENVO:forest_soil	Soil	source	NA	n	5/28/08	Pyrosequencing-Based Assessment of Soil pH as a Predictor of Soil Bacterial Community Structure at the Continental Scale	NA	FA6P1OK	NA	mimarks-survey	NA	NA	NA	98.7	0	NA	"Mary's Peak, OR USA"	1300	2008	NA	0	NA	NA	NA	18.3	8.8	NA	NA	NA	MP2	1	"0.5, g"	NA	NA	targeted gene survey	7.116	NA	-123.5333333	y	0	CCME	NA	NA	NA	"Soils harbor enormously diverse bacterial populations, and soil bacterial communities can vary greatly in composition across space. However, our understanding of the specific changes in soil bacterial community structure that occur across larger spatial scales is limited because most previous work has focused on either surveying a relatively small number of soils in detail or analyzing a larger number of soils with techniques that provide little detail about the phylogenetic structure of the bacterial communities. Here we used a bar-coded pyrosequencing technique to characterize bacterial communities in 88 soils from across North and South America, obtaining an average of 1,501 sequences per soil. We found that overall bacterial community composition, as measured by pairwise UniFrac distances, was significantly correlated with differences in soil pH (r = 0.79), largely driven by changes in the relative abundances of Acidobacteria, Actinobacteria, and Bacteroidetes across the range of soil pHs. In addition, soil pH explains a significant portion of the variability associated with observed changes in the phylogenetic structure within each dominant lineage. The overall phylogenetic diversity of the bacterial communities was also correlated with soil pH (R2 = 0.50), with peak diversity in soils with near-neutral pHs. Together, these results suggest that the structure of soil bacterial communities is predictable, to some degree, across larger spatial scales, and the effect of soil pH on bacterial community composition is evident at even relatively coarse levels of taxonomic resolution."	ENVO:forest soil	TCAG	Pyrosequencing_based_assessment_of_soil_pH_as_a_predictor_of_soil_bacterial_community_structure_at_the_continental_scale	MP2_V2	0	bacterial biogeography	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19502440	NA	NA	NA	Chris Lauber	CCME	GAZ:United States of America	sandy loam	lauber_88_soils	0-0.05	NA	NA	42	NA	n	NA	NA	Lauber_88_soils	pyrosequencing	NA	NA	-1721	soil metagenome	NA	ENVO:Temperate broadleaf and mixed forest biome	FLX	MP2_V2	NA	2200	4.38	NA	Lauber88Soils	Pyrosequencing_based_assessment_of_soil_pH_as_a_predictor_of_soil_bacterial_community_structure_at_the_continental_scale	410658	CCME	inceptisol	NA	NA	103	19502440	lauber_88_soils	NA	y	NA	MP2_V2	ENVO:soil	16S rRNA	NA	NA	V2	CCME	V2	NA	NA	Engencore	NA	49.46666667	NA	NA	soil metagenome
MT1.141719	ACTGTCGAAGCT	CATGCTGCCTCCCGTAGGAGT	chris.lauber@gmail.com	NA	NA	ENVO:forest_soil	Soil	source	NA	n	5/28/08	Pyrosequencing-Based Assessment of Soil pH as a Predictor of Soil Bacterial Community Structure at the Continental Scale	NA	FA6P1OK	NA	mimarks-survey	NA	NA	NA	105.4	0	NA	"Missoula, MT USA"	1000	2008	NA	0	NA	NA	NA	16.007	7	NA	NA	NA	MT1	1	"0.5, g"	NA	NA	targeted gene survey	26.9	NA	-114	y	0	CCME	NA	NA	NA	"Soils harbor enormously diverse bacterial populations, and soil bacterial communities can vary greatly in composition across space. However, our understanding of the specific changes in soil bacterial community structure that occur across larger spatial scales is limited because most previous work has focused on either surveying a relatively small number of soils in detail or analyzing a larger number of soils with techniques that provide little detail about the phylogenetic structure of the bacterial communities. Here we used a bar-coded pyrosequencing technique to characterize bacterial communities in 88 soils from across North and South America, obtaining an average of 1,501 sequences per soil. We found that overall bacterial community composition, as measured by pairwise UniFrac distances, was significantly correlated with differences in soil pH (r = 0.79), largely driven by changes in the relative abundances of Acidobacteria, Actinobacteria, and Bacteroidetes across the range of soil pHs. In addition, soil pH explains a significant portion of the variability associated with observed changes in the phylogenetic structure within each dominant lineage. The overall phylogenetic diversity of the bacterial communities was also correlated with soil pH (R2 = 0.50), with peak diversity in soils with near-neutral pHs. Together, these results suggest that the structure of soil bacterial communities is predictable, to some degree, across larger spatial scales, and the effect of soil pH on bacterial community composition is evident at even relatively coarse levels of taxonomic resolution."	ENVO:forest soil	TCAG	Pyrosequencing_based_assessment_of_soil_pH_as_a_predictor_of_soil_bacterial_community_structure_at_the_continental_scale	MT1_V2	0	bacterial biogeography	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19502440	NA	NA	NA	Chris Lauber	CCME	GAZ:United States of America	sandy loam	lauber_88_soils	0-0.05	NA	NA	30	NA	n	NA	NA	Lauber_88_soils	pyrosequencing	NA	NA	70	soil metagenome	NA	ENVO:forest	FLX	MT1_V2	NA	450	7.57	NA	Lauber88Soils	Pyrosequencing_based_assessment_of_soil_pH_as_a_predictor_of_soil_bacterial_community_structure_at_the_continental_scale	410658	CCME	inceptisol	NA	NA	103	19502440	lauber_88_soils	NA	y	NA	MT1_V2	ENVO:soil	16S rRNA	NA	NA	V2	CCME	V2	NA	NA	Engencore	NA	46.8	NA	NA	soil metagenome
MT2.141698	ACTGTGACTTCA	CATGCTGCCTCCCGTAGGAGT	chris.lauber@gmail.com	NA	NA	ENVO:forest_soil	Soil	source	NA	n	5/28/08	Pyrosequencing-Based Assessment of Soil pH as a Predictor of Soil Bacterial Community Structure at the Continental Scale	NA	FA6P1OK	NA	mimarks-survey	NA	NA	NA	39.1	0	NA	"Missoula, MT USA"	1000	2008	NA	0	NA	NA	NA	23.087	7	NA	NA	NA	MT2	1	"0.5, g"	NA	NA	targeted gene survey	19.7	NA	-114	y	0	CCME	NA	NA	NA	"Soils harbor enormously diverse bacterial populations, and soil bacterial communities can vary greatly in composition across space. However, our understanding of the specific changes in soil bacterial community structure that occur across larger spatial scales is limited because most previous work has focused on either surveying a relatively small number of soils in detail or analyzing a larger number of soils with techniques that provide little detail about the phylogenetic structure of the bacterial communities. Here we used a bar-coded pyrosequencing technique to characterize bacterial communities in 88 soils from across North and South America, obtaining an average of 1,501 sequences per soil. We found that overall bacterial community composition, as measured by pairwise UniFrac distances, was significantly correlated with differences in soil pH (r = 0.79), largely driven by changes in the relative abundances of Acidobacteria, Actinobacteria, and Bacteroidetes across the range of soil pHs. In addition, soil pH explains a significant portion of the variability associated with observed changes in the phylogenetic structure within each dominant lineage. The overall phylogenetic diversity of the bacterial communities was also correlated with soil pH (R2 = 0.50), with peak diversity in soils with near-neutral pHs. Together, these results suggest that the structure of soil bacterial communities is predictable, to some degree, across larger spatial scales, and the effect of soil pH on bacterial community composition is evident at even relatively coarse levels of taxonomic resolution."	ENVO:forest soil	TCAG	Pyrosequencing_based_assessment_of_soil_pH_as_a_predictor_of_soil_bacterial_community_structure_at_the_continental_scale	MT2_V2	0	bacterial biogeography	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19502440	NA	NA	NA	Chris Lauber	CCME	GAZ:United States of America	sandy loam	lauber_88_soils	0-0.05	NA	NA	35	NA	n	NA	NA	Lauber_88_soils	pyrosequencing	NA	NA	70	soil metagenome	NA	ENVO:forest	FLX	MT2_V2	NA	450	6.66	NA	Lauber88Soils	Pyrosequencing_based_assessment_of_soil_pH_as_a_predictor_of_soil_bacterial_community_structure_at_the_continental_scale	410658	CCME	inceptisol	NA	NA	103	19502440	lauber_88_soils	NA	y	NA	MT2_V2	ENVO:soil	16S rRNA	NA	NA	V2	CCME	V2	NA	NA	Engencore	NA	46.8	NA	NA	soil metagenome
SA1.141670	AGATACACGCGC	CATGCTGCCTCCCGTAGGAGT	chris.lauber@gmail.com	NA	NA	ENVO:forest_soil	Soil	source	NA	n	5/28/08	Pyrosequencing-Based Assessment of Soil pH as a Predictor of Soil Bacterial Community Structure at the Continental Scale	NA	FA6P1OK	NA	mimarks-survey	NA	NA	NA	22.9	0	NA	"Sunset Crater, AZ USA"	1905	2008	NA	0	NA	NA	NA	18.2	10.3	NA	NA	NA	SA1	1	"0.5, g"	NA	NA	targeted gene survey	3.821	NA	-111.55	y	0	CCME	NA	NA	NA	"Soils harbor enormously diverse bacterial populations, and soil bacterial communities can vary greatly in composition across space. However, our understanding of the specific changes in soil bacterial community structure that occur across larger spatial scales is limited because most previous work has focused on either surveying a relatively small number of soils in detail or analyzing a larger number of soils with techniques that provide little detail about the phylogenetic structure of the bacterial communities. Here we used a bar-coded pyrosequencing technique to characterize bacterial communities in 88 soils from across North and South America, obtaining an average of 1,501 sequences per soil. We found that overall bacterial community composition, as measured by pairwise UniFrac distances, was significantly correlated with differences in soil pH (r = 0.79), largely driven by changes in the relative abundances of Acidobacteria, Actinobacteria, and Bacteroidetes across the range of soil pHs. In addition, soil pH explains a significant portion of the variability associated with observed changes in the phylogenetic structure within each dominant lineage. The overall phylogenetic diversity of the bacterial communities was also correlated with soil pH (R2 = 0.50), with peak diversity in soils with near-neutral pHs. Together, these results suggest that the structure of soil bacterial communities is predictable, to some degree, across larger spatial scales, and the effect of soil pH on bacterial community composition is evident at even relatively coarse levels of taxonomic resolution."	ENVO:forest soil	TCAG	Pyrosequencing_based_assessment_of_soil_pH_as_a_predictor_of_soil_bacterial_community_structure_at_the_continental_scale	SA1_V2	0	bacterial biogeography	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19502440	NA	NA	NA	Chris Lauber	CCME	GAZ:United States of America	sand	lauber_88_soils	0-0.05	NA	NA	9	NA	n	NA	NA	Lauber_88_soils	pyrosequencing	NA	NA	198	soil metagenome	NA	ENVO:forest	FLX	SA1_V2	NA	400	6.9	NA	Lauber88Soils	Pyrosequencing_based_assessment_of_soil_pH_as_a_predictor_of_soil_bacterial_community_structure_at_the_continental_scale	410658	CCME	entisol	NA	NA	103	19502440	lauber_88_soils	NA	y	NA	SA1_V2	ENVO:soil	16S rRNA	NA	NA	V2	CCME	V2	NA	NA	Engencore	NA	35.36666667	NA	NA	soil metagenome
SK1.141669	AGCACGAGCCTA	CATGCTGCCTCCCGTAGGAGT	chris.lauber@gmail.com	NA	NA	ENVO:forest_soil	Soil	source	NA	n	5/28/08	Pyrosequencing-Based Assessment of Soil pH as a Predictor of Soil Bacterial Community Structure at the Continental Scale	NA	FA6P1OK	NA	mimarks-survey	NA	NA	NA	11.6	0	NA	"BOREAS site, Saskatchewan, Canada"	579	2008	NA	0	NA	NA	NA	31.6767	0.4	NA	NA	NA	SK1	1	"0.5, g"	NA	NA	targeted gene survey	5.8	NA	-104.7	y	0	CCME	NA	NA	NA	"Soils harbor enormously diverse bacterial populations, and soil bacterial communities can vary greatly in composition across space. However, our understanding of the specific changes in soil bacterial community structure that occur across larger spatial scales is limited because most previous work has focused on either surveying a relatively small number of soils in detail or analyzing a larger number of soils with techniques that provide little detail about the phylogenetic structure of the bacterial communities. Here we used a bar-coded pyrosequencing technique to characterize bacterial communities in 88 soils from across North and South America, obtaining an average of 1,501 sequences per soil. We found that overall bacterial community composition, as measured by pairwise UniFrac distances, was significantly correlated with differences in soil pH (r = 0.79), largely driven by changes in the relative abundances of Acidobacteria, Actinobacteria, and Bacteroidetes across the range of soil pHs. In addition, soil pH explains a significant portion of the variability associated with observed changes in the phylogenetic structure within each dominant lineage. The overall phylogenetic diversity of the bacterial communities was also correlated with soil pH (R2 = 0.50), with peak diversity in soils with near-neutral pHs. Together, these results suggest that the structure of soil bacterial communities is predictable, to some degree, across larger spatial scales, and the effect of soil pH on bacterial community composition is evident at even relatively coarse levels of taxonomic resolution."	ENVO:forest soil	TCAG	Pyrosequencing_based_assessment_of_soil_pH_as_a_predictor_of_soil_bacterial_community_structure_at_the_continental_scale	SK1_V2	0	bacterial biogeography	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19502440	NA	NA	NA	Chris Lauber	CCME	GAZ:United States of America	loamy sand	lauber_88_soils	0-0.05	NA	NA	21	NA	n	NA	NA	Lauber_88_soils	pyrosequencing	NA	NA	-13	soil metagenome	NA	ENVO:forest	FLX	SK1_V2	NA	467	5.45	NA	Lauber88Soils	Pyrosequencing_based_assessment_of_soil_pH_as_a_predictor_of_soil_bacterial_community_structure_at_the_continental_scale	410658	CCME	mollisol	NA	NA	103	19502440	lauber_88_soils	NA	y	NA	SK1_V2	ENVO:soil	16S rRNA	NA	NA	V2	CCME	V2	NA	NA	Engencore	NA	53.9	NA	NA	soil metagenome
SK2.141662	AGCAGCACTTGT	CATGCTGCCTCCCGTAGGAGT	chris.lauber@gmail.com	NA	NA	ENVO:forest_soil	Soil	source	NA	n	5/28/08	Pyrosequencing-Based Assessment of Soil pH as a Predictor of Soil Bacterial Community Structure at the Continental Scale	NA	FA6P1OK	NA	mimarks-survey	NA	NA	NA	24.9	0	NA	"BOREAS site, Saskatchewan, Canada"	601	2008	NA	0	NA	NA	NA	24.345	0.4	NA	NA	NA	SK2	1	"0.5, g"	NA	NA	targeted gene survey	6.5	NA	-105.2	y	0	CCME	NA	NA	NA	"Soils harbor enormously diverse bacterial populations, and soil bacterial communities can vary greatly in composition across space. However, our understanding of the specific changes in soil bacterial community structure that occur across larger spatial scales is limited because most previous work has focused on either surveying a relatively small number of soils in detail or analyzing a larger number of soils with techniques that provide little detail about the phylogenetic structure of the bacterial communities. Here we used a bar-coded pyrosequencing technique to characterize bacterial communities in 88 soils from across North and South America, obtaining an average of 1,501 sequences per soil. We found that overall bacterial community composition, as measured by pairwise UniFrac distances, was significantly correlated with differences in soil pH (r = 0.79), largely driven by changes in the relative abundances of Acidobacteria, Actinobacteria, and Bacteroidetes across the range of soil pHs. In addition, soil pH explains a significant portion of the variability associated with observed changes in the phylogenetic structure within each dominant lineage. The overall phylogenetic diversity of the bacterial communities was also correlated with soil pH (R2 = 0.50), with peak diversity in soils with near-neutral pHs. Together, these results suggest that the structure of soil bacterial communities is predictable, to some degree, across larger spatial scales, and the effect of soil pH on bacterial community composition is evident at even relatively coarse levels of taxonomic resolution."	ENVO:forest soil	TCAG	Pyrosequencing_based_assessment_of_soil_pH_as_a_predictor_of_soil_bacterial_community_structure_at_the_continental_scale	SK2_V2	0	bacterial biogeography	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19502440	NA	NA	NA	Chris Lauber	CCME	GAZ:United States of America	sandy loam	lauber_88_soils	0-0.05	NA	NA	35	NA	n	NA	NA	Lauber_88_soils	pyrosequencing	NA	NA	-13	soil metagenome	NA	ENVO:forest	FLX	SK2_V2	NA	467	6.18	NA	Lauber88Soils	Pyrosequencing_based_assessment_of_soil_pH_as_a_predictor_of_soil_bacterial_community_structure_at_the_continental_scale	410658	CCME	mollisol	NA	NA	103	19502440	lauber_88_soils	NA	y	NA	SK2_V2	ENVO:soil	16S rRNA	NA	NA	V2	CCME	V2	NA	NA	Engencore	NA	53.98333333	NA	NA	soil metagenome
SK3.141716	AGCAGTCGCGAT	CATGCTGCCTCCCGTAGGAGT	chris.lauber@gmail.com	NA	NA	ENVO:forest_soil	Soil	source	NA	n	5/28/08	Pyrosequencing-Based Assessment of Soil pH as a Predictor of Soil Bacterial Community Structure at the Continental Scale	NA	FA6P1OK	NA	mimarks-survey	NA	NA	NA	8.5	0	NA	"BOREAS site, Saskatchewan, Canada"	601	2008	NA	0	NA	NA	NA	14.135	0.4	NA	NA	NA	SK3	1	"0.5, g"	NA	NA	targeted gene survey	3.7	NA	-106.2	y	0	CCME	NA	NA	NA	"Soils harbor enormously diverse bacterial populations, and soil bacterial communities can vary greatly in composition across space. However, our understanding of the specific changes in soil bacterial community structure that occur across larger spatial scales is limited because most previous work has focused on either surveying a relatively small number of soils in detail or analyzing a larger number of soils with techniques that provide little detail about the phylogenetic structure of the bacterial communities. Here we used a bar-coded pyrosequencing technique to characterize bacterial communities in 88 soils from across North and South America, obtaining an average of 1,501 sequences per soil. We found that overall bacterial community composition, as measured by pairwise UniFrac distances, was significantly correlated with differences in soil pH (r = 0.79), largely driven by changes in the relative abundances of Acidobacteria, Actinobacteria, and Bacteroidetes across the range of soil pHs. In addition, soil pH explains a significant portion of the variability associated with observed changes in the phylogenetic structure within each dominant lineage. The overall phylogenetic diversity of the bacterial communities was also correlated with soil pH (R2 = 0.50), with peak diversity in soils with near-neutral pHs. Together, these results suggest that the structure of soil bacterial communities is predictable, to some degree, across larger spatial scales, and the effect of soil pH on bacterial community composition is evident at even relatively coarse levels of taxonomic resolution."	ENVO:forest soil	TCAG	Pyrosequencing_based_assessment_of_soil_pH_as_a_predictor_of_soil_bacterial_community_structure_at_the_continental_scale	SK3_V2	0	bacterial biogeography	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19502440	NA	NA	NA	Chris Lauber	CCME	GAZ:United States of America	loam	lauber_88_soils	0-0.05	NA	NA	55	NA	n	NA	NA	Lauber_88_soils	pyrosequencing	NA	NA	-13	soil metagenome	NA	ENVO:forest	FLX	SK3_V2	NA	467	5.83	NA	Lauber88Soils	Pyrosequencing_based_assessment_of_soil_pH_as_a_predictor_of_soil_bacterial_community_structure_at_the_continental_scale	410658	CCME	mollisol	NA	NA	103	19502440	lauber_88_soils	NA	y	NA	SK3_V2	ENVO:soil	16S rRNA	NA	NA	V2	CCME	V2	NA	NA	Engencore	NA	53.6	NA	NA	soil metagenome
SN1.141681	AGCATATGAGAG	CATGCTGCCTCCCGTAGGAGT	chris.lauber@gmail.com	NA	NA	ENVO:forest_soil	Soil	source	NA	n	5/28/08	Pyrosequencing-Based Assessment of Soil pH as a Predictor of Soil Bacterial Community Structure at the Continental Scale	NA	FA6P1OK	NA	mimarks-survey	NA	NA	NA	42.5	0	NA	"Sierra Nevada Mts., CA USA"	3000	2008	NA	0	NA	NA	NA	22	3.6	NA	NA	NA	SN1	1	"0.5, g"	NA	NA	targeted gene survey	12.996	NA	-118.1666667	y	0	CCME	NA	NA	NA	"Soils harbor enormously diverse bacterial populations, and soil bacterial communities can vary greatly in composition across space. However, our understanding of the specific changes in soil bacterial community structure that occur across larger spatial scales is limited because most previous work has focused on either surveying a relatively small number of soils in detail or analyzing a larger number of soils with techniques that provide little detail about the phylogenetic structure of the bacterial communities. Here we used a bar-coded pyrosequencing technique to characterize bacterial communities in 88 soils from across North and South America, obtaining an average of 1,501 sequences per soil. We found that overall bacterial community composition, as measured by pairwise UniFrac distances, was significantly correlated with differences in soil pH (r = 0.79), largely driven by changes in the relative abundances of Acidobacteria, Actinobacteria, and Bacteroidetes across the range of soil pHs. In addition, soil pH explains a significant portion of the variability associated with observed changes in the phylogenetic structure within each dominant lineage. The overall phylogenetic diversity of the bacterial communities was also correlated with soil pH (R2 = 0.50), with peak diversity in soils with near-neutral pHs. Together, these results suggest that the structure of soil bacterial communities is predictable, to some degree, across larger spatial scales, and the effect of soil pH on bacterial community composition is evident at even relatively coarse levels of taxonomic resolution."	ENVO:forest soil	TCAG	Pyrosequencing_based_assessment_of_soil_pH_as_a_predictor_of_soil_bacterial_community_structure_at_the_continental_scale	SN1_V2	0	bacterial biogeography	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19502440	NA	NA	NA	Chris Lauber	CCME	GAZ:United States of America	loamy sand	lauber_88_soils	0-0.05	NA	NA	20	NA	n	NA	NA	Lauber_88_soils	pyrosequencing	NA	NA	-252	soil metagenome	NA	ENVO:Temperate broadleaf and mixed forest biome	FLX	SN1_V2	NA	600	4.95	NA	Lauber88Soils	Pyrosequencing_based_assessment_of_soil_pH_as_a_predictor_of_soil_bacterial_community_structure_at_the_continental_scale	410658	CCME	inceptisol	NA	NA	103	19502440	lauber_88_soils	NA	y	NA	SN1_V2	ENVO:soil	16S rRNA	NA	NA	V2	CCME	V2	NA	NA	Engencore	NA	36.45	NA	NA	soil metagenome
VC1.141722	AGGTGTGATCGC	CATGCTGCCTCCCGTAGGAGT	chris.lauber@gmail.com	NA	NA	ENVO:forest_soil	Soil	source	NA	n	5/28/08	Pyrosequencing-Based Assessment of Soil pH as a Predictor of Soil Bacterial Community Structure at the Continental Scale	NA	FA6P1OK	NA	mimarks-survey	NA	NA	NA	56.7	0	NA	"Valles Caldera, NM USA"	2746	2008	NA	0	NA	NA	NA	17	2.5	NA	NA	NA	VC1	1	"0.5, g"	NA	NA	targeted gene survey	15.209	NA	-106.55	y	0	CCME	NA	NA	NA	"Soils harbor enormously diverse bacterial populations, and soil bacterial communities can vary greatly in composition across space. However, our understanding of the specific changes in soil bacterial community structure that occur across larger spatial scales is limited because most previous work has focused on either surveying a relatively small number of soils in detail or analyzing a larger number of soils with techniques that provide little detail about the phylogenetic structure of the bacterial communities. Here we used a bar-coded pyrosequencing technique to characterize bacterial communities in 88 soils from across North and South America, obtaining an average of 1,501 sequences per soil. We found that overall bacterial community composition, as measured by pairwise UniFrac distances, was significantly correlated with differences in soil pH (r = 0.79), largely driven by changes in the relative abundances of Acidobacteria, Actinobacteria, and Bacteroidetes across the range of soil pHs. In addition, soil pH explains a significant portion of the variability associated with observed changes in the phylogenetic structure within each dominant lineage. The overall phylogenetic diversity of the bacterial communities was also correlated with soil pH (R2 = 0.50), with peak diversity in soils with near-neutral pHs. Together, these results suggest that the structure of soil bacterial communities is predictable, to some degree, across larger spatial scales, and the effect of soil pH on bacterial community composition is evident at even relatively coarse levels of taxonomic resolution."	ENVO:forest soil	TCAG	Pyrosequencing_based_assessment_of_soil_pH_as_a_predictor_of_soil_bacterial_community_structure_at_the_continental_scale	VC1_V2	0	bacterial biogeography	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19502440	NA	NA	NA	Chris Lauber	CCME	GAZ:United States of America	sandy loam	lauber_88_soils	0-0.05	NA	NA	41	NA	n	NA	NA	Lauber_88_soils	pyrosequencing	NA	NA	-175	soil metagenome	NA	ENVO:forest	FLX	VC1_V2	NA	500	5.55	NA	Lauber88Soils	Pyrosequencing_based_assessment_of_soil_pH_as_a_predictor_of_soil_bacterial_community_structure_at_the_continental_scale	410658	CCME	mollisol	NA	NA	103	19502440	lauber_88_soils	NA	y	NA	VC1_V2	ENVO:soil	16S rRNA	NA	NA	V2	CCME	V2	NA	NA	Engencore	NA	35.9	NA	NA	soil metagenome
BP1.141702	ACACACTATGGC	CATGCTGCCTCCCGTAGGAGT	chris.lauber@gmail.com	NA	NA	ENVO:grassland_soil	Soil	source	NA	n	5/28/08	Pyrosequencing-Based Assessment of Soil pH as a Predictor of Soil Bacterial Community Structure at the Continental Scale	NA	FA6P1OK	NA	mimarks-survey	NA	NA	NA	31	0	NA	"Badlands National Park, SD USA"	1100	2008	NA	0	NA	NA	NA	14.5	6.6	NA	NA	NA	BP1	1	"0.5, g"	NA	NA	targeted gene survey	7.736	NA	-102.3833333	y	0	CCME	NA	NA	NA	"Soils harbor enormously diverse bacterial populations, and soil bacterial communities can vary greatly in composition across space. However, our understanding of the specific changes in soil bacterial community structure that occur across larger spatial scales is limited because most previous work has focused on either surveying a relatively small number of soils in detail or analyzing a larger number of soils with techniques that provide little detail about the phylogenetic structure of the bacterial communities. Here we used a bar-coded pyrosequencing technique to characterize bacterial communities in 88 soils from across North and South America, obtaining an average of 1,501 sequences per soil. We found that overall bacterial community composition, as measured by pairwise UniFrac distances, was significantly correlated with differences in soil pH (r = 0.79), largely driven by changes in the relative abundances of Acidobacteria, Actinobacteria, and Bacteroidetes across the range of soil pHs. In addition, soil pH explains a significant portion of the variability associated with observed changes in the phylogenetic structure within each dominant lineage. The overall phylogenetic diversity of the bacterial communities was also correlated with soil pH (R2 = 0.50), with peak diversity in soils with near-neutral pHs. Together, these results suggest that the structure of soil bacterial communities is predictable, to some degree, across larger spatial scales, and the effect of soil pH on bacterial community composition is evident at even relatively coarse levels of taxonomic resolution."	ENVO:grassland soil	TCAG	Pyrosequencing_based_assessment_of_soil_pH_as_a_predictor_of_soil_bacterial_community_structure_at_the_continental_scale	BP1_V2	0	bacterial biogeography	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19502440	NA	NA	NA	Chris Lauber	CCME	GAZ:United States of America	silt loam	lauber_88_soils	0-0.05	NA	NA	65	NA	n	NA	NA	Lauber_88_soils	pyrosequencing	NA	NA	79	soil metagenome	NA	ENVO:grassland	FLX	BP1_V2	NA	450	7.53	NA	Lauber88Soils	Pyrosequencing_based_assessment_of_soil_pH_as_a_predictor_of_soil_bacterial_community_structure_at_the_continental_scale	410658	CCME	entisol	NA	NA	103	19502440	lauber_88_soils	NA	y	NA	BP1_V2	ENVO:soil	16S rRNA	NA	NA	V2	CCME	V2	NA	NA	Engencore	NA	43.75	NA	NA	soil metagenome
CC1.141721	ACACTAGATCCG	CATGCTGCCTCCCGTAGGAGT	chris.lauber@gmail.com	NA	NA	ENVO:grassland_soil	Soil	source	NA	n	5/28/08	Pyrosequencing-Based Assessment of Soil pH as a Predictor of Soil Bacterial Community Structure at the Continental Scale	NA	FA6P1OK	NA	mimarks-survey	NA	NA	NA	19.1	0	NA	"Cedar Creek LTER, MN USA"	110	2008	NA	0	NA	NA	NA	14.5	5.8	NA	NA	NA	CC1	1	"0.5, g"	NA	NA	targeted gene survey	4.204	NA	-93.2	y	0	CCME	NA	NA	NA	"Soils harbor enormously diverse bacterial populations, and soil bacterial communities can vary greatly in composition across space. However, our understanding of the specific changes in soil bacterial community structure that occur across larger spatial scales is limited because most previous work has focused on either surveying a relatively small number of soils in detail or analyzing a larger number of soils with techniques that provide little detail about the phylogenetic structure of the bacterial communities. Here we used a bar-coded pyrosequencing technique to characterize bacterial communities in 88 soils from across North and South America, obtaining an average of 1,501 sequences per soil. We found that overall bacterial community composition, as measured by pairwise UniFrac distances, was significantly correlated with differences in soil pH (r = 0.79), largely driven by changes in the relative abundances of Acidobacteria, Actinobacteria, and Bacteroidetes across the range of soil pHs. In addition, soil pH explains a significant portion of the variability associated with observed changes in the phylogenetic structure within each dominant lineage. The overall phylogenetic diversity of the bacterial communities was also correlated with soil pH (R2 = 0.50), with peak diversity in soils with near-neutral pHs. Together, these results suggest that the structure of soil bacterial communities is predictable, to some degree, across larger spatial scales, and the effect of soil pH on bacterial community composition is evident at even relatively coarse levels of taxonomic resolution."	ENVO:grassland soil	TCAG	Pyrosequencing_based_assessment_of_soil_pH_as_a_predictor_of_soil_bacterial_community_structure_at_the_continental_scale	CC1_V2	0	bacterial biogeography	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19502440	NA	NA	NA	Chris Lauber	CCME	GAZ:United States of America	sand	lauber_88_soils	0-0.05	NA	NA	11	NA	n	NA	NA	Lauber_88_soils	pyrosequencing	NA	NA	-143	soil metagenome	NA	ENVO:Temperate grasslands	FLX	CC1_V2	NA	720	6.06	NA	Lauber88Soils	Pyrosequencing_based_assessment_of_soil_pH_as_a_predictor_of_soil_bacterial_community_structure_at_the_continental_scale	410658	CCME	inceptisol	NA	NA	103	19502440	lauber_88_soils	NA	y	NA	CC1_V2	ENVO:soil	16S rRNA	NA	NA	V2	CCME	V2	NA	NA	Engencore	NA	45.4	NA	NA	soil metagenome
CL2.141671	ACAGCTAGCTTG	CATGCTGCCTCCCGTAGGAGT	chris.lauber@gmail.com	NA	NA	ENVO:grassland_soil	Soil	source	NA	n	5/28/08	Pyrosequencing-Based Assessment of Soil pH as a Predictor of Soil Bacterial Community Structure at the Continental Scale	NA	FA6P1OK	NA	mimarks-survey	NA	NA	NA	22.7	0	NA	"Calhoun Experimental Forest, SC USA"	150	2008	NA	0	NA	NA	NA	12.8	15.9	NA	NA	NA	CL2	1	"0.5, g"	NA	NA	targeted gene survey	4.07	NA	-81.66666667	y	0	CCME	NA	NA	NA	"Soils harbor enormously diverse bacterial populations, and soil bacterial communities can vary greatly in composition across space. However, our understanding of the specific changes in soil bacterial community structure that occur across larger spatial scales is limited because most previous work has focused on either surveying a relatively small number of soils in detail or analyzing a larger number of soils with techniques that provide little detail about the phylogenetic structure of the bacterial communities. Here we used a bar-coded pyrosequencing technique to characterize bacterial communities in 88 soils from across North and South America, obtaining an average of 1,501 sequences per soil. We found that overall bacterial community composition, as measured by pairwise UniFrac distances, was significantly correlated with differences in soil pH (r = 0.79), largely driven by changes in the relative abundances of Acidobacteria, Actinobacteria, and Bacteroidetes across the range of soil pHs. In addition, soil pH explains a significant portion of the variability associated with observed changes in the phylogenetic structure within each dominant lineage. The overall phylogenetic diversity of the bacterial communities was also correlated with soil pH (R2 = 0.50), with peak diversity in soils with near-neutral pHs. Together, these results suggest that the structure of soil bacterial communities is predictable, to some degree, across larger spatial scales, and the effect of soil pH on bacterial community composition is evident at even relatively coarse levels of taxonomic resolution."	ENVO:grassland soil	TCAG	Pyrosequencing_based_assessment_of_soil_pH_as_a_predictor_of_soil_bacterial_community_structure_at_the_continental_scale	CL2_V2	0	bacterial biogeography	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19502440	NA	NA	NA	Chris Lauber	CCME	GAZ:United States of America	sandy loam	lauber_88_soils	0-0.05	NA	NA	25	NA	n	NA	NA	Lauber_88_soils	pyrosequencing	NA	NA	-420	soil metagenome	NA	ENVO:Temperate grasslands	FLX	CL2_V2	NA	1250	5.57	NA	Lauber88Soils	Pyrosequencing_based_assessment_of_soil_pH_as_a_predictor_of_soil_bacterial_community_structure_at_the_continental_scale	410658	CCME	ultisol	NA	NA	103	19502440	lauber_88_soils	NA	y	NA	CL2_V2	ENVO:soil	16S rRNA	NA	NA	V2	CCME	V2	NA	NA	Engencore	NA	34.61666667	NA	NA	soil metagenome
CL4.141667	ACAGTTGCGCGA	CATGCTGCCTCCCGTAGGAGT	chris.lauber@gmail.com	NA	NA	ENVO:grassland_soil	Soil	source	NA	n	5/28/08	Pyrosequencing-Based Assessment of Soil pH as a Predictor of Soil Bacterial Community Structure at the Continental Scale	NA	FA6P1OK	NA	mimarks-survey	NA	NA	NA	17.1	0	NA	"Calhoun Experimental Forest, SC USA"	150	2008	NA	0	NA	NA	NA	13.7	15.9	NA	NA	NA	CL4	1	"0.5, g"	NA	NA	targeted gene survey	2.234	NA	-81.66666667	y	0	CCME	NA	NA	NA	"Soils harbor enormously diverse bacterial populations, and soil bacterial communities can vary greatly in composition across space. However, our understanding of the specific changes in soil bacterial community structure that occur across larger spatial scales is limited because most previous work has focused on either surveying a relatively small number of soils in detail or analyzing a larger number of soils with techniques that provide little detail about the phylogenetic structure of the bacterial communities. Here we used a bar-coded pyrosequencing technique to characterize bacterial communities in 88 soils from across North and South America, obtaining an average of 1,501 sequences per soil. We found that overall bacterial community composition, as measured by pairwise UniFrac distances, was significantly correlated with differences in soil pH (r = 0.79), largely driven by changes in the relative abundances of Acidobacteria, Actinobacteria, and Bacteroidetes across the range of soil pHs. In addition, soil pH explains a significant portion of the variability associated with observed changes in the phylogenetic structure within each dominant lineage. The overall phylogenetic diversity of the bacterial communities was also correlated with soil pH (R2 = 0.50), with peak diversity in soils with near-neutral pHs. Together, these results suggest that the structure of soil bacterial communities is predictable, to some degree, across larger spatial scales, and the effect of soil pH on bacterial community composition is evident at even relatively coarse levels of taxonomic resolution."	ENVO:grassland soil	TCAG	Pyrosequencing_based_assessment_of_soil_pH_as_a_predictor_of_soil_bacterial_community_structure_at_the_continental_scale	CL4_V2	0	bacterial biogeography	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19502440	NA	NA	NA	Chris Lauber	CCME	GAZ:United States of America	sandy loam	lauber_88_soils	0-0.05	NA	NA	36	NA	n	NA	NA	Lauber_88_soils	pyrosequencing	NA	NA	-420	soil metagenome	NA	ENVO:Temperate grasslands	FLX	CL4_V2	NA	1250	5.03	NA	Lauber88Soils	Pyrosequencing_based_assessment_of_soil_pH_as_a_predictor_of_soil_bacterial_community_structure_at_the_continental_scale	410658	CCME	ultisol	NA	NA	103	19502440	lauber_88_soils	NA	y	NA	CL4_V2	ENVO:soil	16S rRNA	NA	NA	V2	CCME	V2	NA	NA	Engencore	NA	34.61666667	NA	NA	soil metagenome
CM1.141723	ACATCACTTAGC	CATGCTGCCTCCCGTAGGAGT	chris.lauber@gmail.com	NA	NA	ENVO:grassland_soil	Soil	source	NA	n	5/28/08	Pyrosequencing-Based Assessment of Soil pH as a Predictor of Soil Bacterial Community Structure at the Continental Scale	NA	FA6P1OK	NA	mimarks-survey	NA	NA	NA	29.9	0	NA	"Clymer Meadow Preserve, TX USA"	200	2008	NA	0	NA	NA	NA	12.4	18.5	NA	NA	NA	CM1	1	"0.5, g"	NA	NA	targeted gene survey	2.043	NA	-96.23333333	y	10.2	CCME	NA	NA	NA	"Soils harbor enormously diverse bacterial populations, and soil bacterial communities can vary greatly in composition across space. However, our understanding of the specific changes in soil bacterial community structure that occur across larger spatial scales is limited because most previous work has focused on either surveying a relatively small number of soils in detail or analyzing a larger number of soils with techniques that provide little detail about the phylogenetic structure of the bacterial communities. Here we used a bar-coded pyrosequencing technique to characterize bacterial communities in 88 soils from across North and South America, obtaining an average of 1,501 sequences per soil. We found that overall bacterial community composition, as measured by pairwise UniFrac distances, was significantly correlated with differences in soil pH (r = 0.79), largely driven by changes in the relative abundances of Acidobacteria, Actinobacteria, and Bacteroidetes across the range of soil pHs. In addition, soil pH explains a significant portion of the variability associated with observed changes in the phylogenetic structure within each dominant lineage. The overall phylogenetic diversity of the bacterial communities was also correlated with soil pH (R2 = 0.50), with peak diversity in soils with near-neutral pHs. Together, these results suggest that the structure of soil bacterial communities is predictable, to some degree, across larger spatial scales, and the effect of soil pH on bacterial community composition is evident at even relatively coarse levels of taxonomic resolution."	ENVO:grassland soil	TCAG	Pyrosequencing_based_assessment_of_soil_pH_as_a_predictor_of_soil_bacterial_community_structure_at_the_continental_scale	CM1_V2	0	bacterial biogeography	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19502440	NA	NA	NA	Chris Lauber	CCME	GAZ:United States of America	silty clay	lauber_88_soils	0-0.05	NA	NA	82	NA	n	NA	NA	Lauber_88_soils	pyrosequencing	NA	NA	155	soil metagenome	NA	ENVO:Temperate grasslands	FLX	CM1_V2	NA	850	7.85	NA	Lauber88Soils	Pyrosequencing_based_assessment_of_soil_pH_as_a_predictor_of_soil_bacterial_community_structure_at_the_continental_scale	410658	CCME	mollisol	NA	NA	103	19502440	lauber_88_soils	NA	y	NA	CM1_V2	ENVO:soil	16S rRNA	NA	NA	V2	CCME	V2	NA	NA	Engencore	NA	33.3	NA	NA	soil metagenome
CO3.141651	ACATTCAGCGCA	CATGCTGCCTCCCGTAGGAGT	chris.lauber@gmail.com	NA	NA	ENVO:grassland_soil	Soil	source	NA	n	5/28/08	Pyrosequencing-Based Assessment of Soil pH as a Predictor of Soil Bacterial Community Structure at the Continental Scale	NA	FA6P1OK	NA	mimarks-survey	NA	NA	NA	8.2	0	NA	"Shortgrass Steppe LTER, CO USA"	1500	2008	NA	0	NA	NA	NA	11.7	9.3	NA	NA	NA	CO3	1	"0.5, g"	NA	NA	targeted gene survey	2.443	NA	-104.8333333	y	0	CCME	NA	NA	NA	"Soils harbor enormously diverse bacterial populations, and soil bacterial communities can vary greatly in composition across space. However, our understanding of the specific changes in soil bacterial community structure that occur across larger spatial scales is limited because most previous work has focused on either surveying a relatively small number of soils in detail or analyzing a larger number of soils with techniques that provide little detail about the phylogenetic structure of the bacterial communities. Here we used a bar-coded pyrosequencing technique to characterize bacterial communities in 88 soils from across North and South America, obtaining an average of 1,501 sequences per soil. We found that overall bacterial community composition, as measured by pairwise UniFrac distances, was significantly correlated with differences in soil pH (r = 0.79), largely driven by changes in the relative abundances of Acidobacteria, Actinobacteria, and Bacteroidetes across the range of soil pHs. In addition, soil pH explains a significant portion of the variability associated with observed changes in the phylogenetic structure within each dominant lineage. The overall phylogenetic diversity of the bacterial communities was also correlated with soil pH (R2 = 0.50), with peak diversity in soils with near-neutral pHs. Together, these results suggest that the structure of soil bacterial communities is predictable, to some degree, across larger spatial scales, and the effect of soil pH on bacterial community composition is evident at even relatively coarse levels of taxonomic resolution."	ENVO:grassland soil	TCAG	Pyrosequencing_based_assessment_of_soil_pH_as_a_predictor_of_soil_bacterial_community_structure_at_the_continental_scale	CO3_V2	0	bacterial biogeography	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19502440	NA	NA	NA	Chris Lauber	CCME	GAZ:United States of America	sandy loam	lauber_88_soils	0-0.05	NA	NA	24	NA	n	NA	NA	Lauber_88_soils	pyrosequencing	NA	NA	231	soil metagenome	NA	ENVO:grassland	FLX	CO3_V2	NA	322	6.02	NA	Lauber88Soils	Pyrosequencing_based_assessment_of_soil_pH_as_a_predictor_of_soil_bacterial_community_structure_at_the_continental_scale	410658	CCME	mollisol	NA	NA	103	19502440	lauber_88_soils	NA	y	NA	CO3_V2	ENVO:soil	16S rRNA	NA	NA	V2	CCME	V2	NA	NA	Engencore	NA	40.8	NA	NA	soil metagenome
CR1.141682	ACCACATACATC	CATGCTGCCTCCCGTAGGAGT	chris.lauber@gmail.com	NA	NA	ENVO:grassland_soil	Soil	source	NA	n	5/28/08	Pyrosequencing-Based Assessment of Soil pH as a Predictor of Soil Bacterial Community Structure at the Continental Scale	NA	FA6P1OK	NA	mimarks-survey	NA	NA	NA	28.3	0	NA	"Coffey Ranch, TX USA"	250	2008	NA	0	NA	NA	NA	10.7	18.4	NA	NA	NA	CR1	1	"0.5, g"	NA	NA	targeted gene survey	3.078	NA	-97.23333333	y	10.9	CCME	NA	NA	NA	"Soils harbor enormously diverse bacterial populations, and soil bacterial communities can vary greatly in composition across space. However, our understanding of the specific changes in soil bacterial community structure that occur across larger spatial scales is limited because most previous work has focused on either surveying a relatively small number of soils in detail or analyzing a larger number of soils with techniques that provide little detail about the phylogenetic structure of the bacterial communities. Here we used a bar-coded pyrosequencing technique to characterize bacterial communities in 88 soils from across North and South America, obtaining an average of 1,501 sequences per soil. We found that overall bacterial community composition, as measured by pairwise UniFrac distances, was significantly correlated with differences in soil pH (r = 0.79), largely driven by changes in the relative abundances of Acidobacteria, Actinobacteria, and Bacteroidetes across the range of soil pHs. In addition, soil pH explains a significant portion of the variability associated with observed changes in the phylogenetic structure within each dominant lineage. The overall phylogenetic diversity of the bacterial communities was also correlated with soil pH (R2 = 0.50), with peak diversity in soils with near-neutral pHs. Together, these results suggest that the structure of soil bacterial communities is predictable, to some degree, across larger spatial scales, and the effect of soil pH on bacterial community composition is evident at even relatively coarse levels of taxonomic resolution."	ENVO:grassland soil	TCAG	Pyrosequencing_based_assessment_of_soil_pH_as_a_predictor_of_soil_bacterial_community_structure_at_the_continental_scale	CR1_V2	0	bacterial biogeography	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19502440	NA	NA	NA	Chris Lauber	CCME	GAZ:United States of America	loam	lauber_88_soils	0-0.05	NA	NA	64	NA	n	NA	NA	Lauber_88_soils	pyrosequencing	NA	NA	160	soil metagenome	NA	ENVO:Temperate grasslands	FLX	CR1_V2	NA	850	8	NA	Lauber88Soils	Pyrosequencing_based_assessment_of_soil_pH_as_a_predictor_of_soil_bacterial_community_structure_at_the_continental_scale	410658	CCME	mollisol	NA	NA	103	19502440	lauber_88_soils	NA	y	NA	CR1_V2	ENVO:soil	16S rRNA	NA	NA	V2	CCME	V2	NA	NA	Engencore	NA	33.93333333	NA	NA	soil metagenome
GB1.141665	ACCTCGATCAGA	CATGCTGCCTCCCGTAGGAGT	chris.lauber@gmail.com	NA	NA	ENVO:grassland_soil	Soil	source	NA	n	5/28/08	Pyrosequencing-Based Assessment of Soil pH as a Predictor of Soil Bacterial Community Structure at the Continental Scale	NA	FA6P1OK	NA	mimarks-survey	NA	NA	NA	28.2	0	NA	"Great Basin Experimental Range, UT USA"	3750	2008	NA	0	NA	NA	NA	12.2	2	NA	NA	NA	GB1	1	"0.5, g"	NA	NA	targeted gene survey	6.098	NA	-111.45	y	0	CCME	NA	NA	NA	"Soils harbor enormously diverse bacterial populations, and soil bacterial communities can vary greatly in composition across space. However, our understanding of the specific changes in soil bacterial community structure that occur across larger spatial scales is limited because most previous work has focused on either surveying a relatively small number of soils in detail or analyzing a larger number of soils with techniques that provide little detail about the phylogenetic structure of the bacterial communities. Here we used a bar-coded pyrosequencing technique to characterize bacterial communities in 88 soils from across North and South America, obtaining an average of 1,501 sequences per soil. We found that overall bacterial community composition, as measured by pairwise UniFrac distances, was significantly correlated with differences in soil pH (r = 0.79), largely driven by changes in the relative abundances of Acidobacteria, Actinobacteria, and Bacteroidetes across the range of soil pHs. In addition, soil pH explains a significant portion of the variability associated with observed changes in the phylogenetic structure within each dominant lineage. The overall phylogenetic diversity of the bacterial communities was also correlated with soil pH (R2 = 0.50), with peak diversity in soils with near-neutral pHs. Together, these results suggest that the structure of soil bacterial communities is predictable, to some degree, across larger spatial scales, and the effect of soil pH on bacterial community composition is evident at even relatively coarse levels of taxonomic resolution."	ENVO:grassland soil	TCAG	Pyrosequencing_based_assessment_of_soil_pH_as_a_predictor_of_soil_bacterial_community_structure_at_the_continental_scale	GB1_V2	0	bacterial biogeography	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19502440	NA	NA	NA	Chris Lauber	CCME	GAZ:United States of America	clay loam	lauber_88_soils	0-0.05	NA	NA	77	NA	n	NA	NA	Lauber_88_soils	pyrosequencing	NA	NA	-112	soil metagenome	NA	ENVO:grassland	FLX	GB1_V2	NA	400	6.84	NA	Lauber88Soils	Pyrosequencing_based_assessment_of_soil_pH_as_a_predictor_of_soil_bacterial_community_structure_at_the_continental_scale	410658	CCME	mollisol	NA	NA	103	19502440	lauber_88_soils	NA	y	NA	GB1_V2	ENVO:soil	16S rRNA	NA	NA	V2	CCME	V2	NA	NA	Engencore	NA	39.33333333	NA	NA	soil metagenome
HI2.141672	ACGCGCAGATAC	CATGCTGCCTCCCGTAGGAGT	chris.lauber@gmail.com	NA	NA	ENVO:grassland_soil	Soil	source	NA	n	5/28/08	Pyrosequencing-Based Assessment of Soil pH as a Predictor of Soil Bacterial Community Structure at the Continental Scale	NA	FA6P1OK	NA	mimarks-survey	NA	NA	NA	158.8	0	NA	"Kohala Peninsula, HI USA"	1000	2008	NA	0	NA	NA	NA	11	22.8	NA	NA	NA	HI2	1	"0.5, g"	NA	NA	targeted gene survey	11.805	NA	-155.7	y	0	CCME	NA	NA	NA	"Soils harbor enormously diverse bacterial populations, and soil bacterial communities can vary greatly in composition across space. However, our understanding of the specific changes in soil bacterial community structure that occur across larger spatial scales is limited because most previous work has focused on either surveying a relatively small number of soils in detail or analyzing a larger number of soils with techniques that provide little detail about the phylogenetic structure of the bacterial communities. Here we used a bar-coded pyrosequencing technique to characterize bacterial communities in 88 soils from across North and South America, obtaining an average of 1,501 sequences per soil. We found that overall bacterial community composition, as measured by pairwise UniFrac distances, was significantly correlated with differences in soil pH (r = 0.79), largely driven by changes in the relative abundances of Acidobacteria, Actinobacteria, and Bacteroidetes across the range of soil pHs. In addition, soil pH explains a significant portion of the variability associated with observed changes in the phylogenetic structure within each dominant lineage. The overall phylogenetic diversity of the bacterial communities was also correlated with soil pH (R2 = 0.50), with peak diversity in soils with near-neutral pHs. Together, these results suggest that the structure of soil bacterial communities is predictable, to some degree, across larger spatial scales, and the effect of soil pH on bacterial community composition is evident at even relatively coarse levels of taxonomic resolution."	ENVO:grassland soil	TCAG	Pyrosequencing_based_assessment_of_soil_pH_as_a_predictor_of_soil_bacterial_community_structure_at_the_continental_scale	HI2_V2	0	bacterial biogeography	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19502440	NA	NA	NA	Chris Lauber	CCME	GAZ:United States of America	sandy loam	lauber_88_soils	0-0.05	NA	NA	36	NA	n	NA	NA	Lauber_88_soils	pyrosequencing	NA	NA	358	soil metagenome	NA	"ENVO:Tropical and subtropical grasslands, savannas, and shrubland biome"	FLX	HI2_V2	NA	750	6.32	NA	Lauber88Soils	Pyrosequencing_based_assessment_of_soil_pH_as_a_predictor_of_soil_bacterial_community_structure_at_the_continental_scale	410658	CCME	andisol	NA	NA	103	19502440	lauber_88_soils	NA	y	NA	HI2_V2	ENVO:soil	16S rRNA	NA	NA	V2	CCME	V2	NA	NA	Engencore	NA	20.08333333	NA	NA	soil metagenome
HI3.141676	ACGCTATCTGGA	CATGCTGCCTCCCGTAGGAGT	chris.lauber@gmail.com	NA	NA	ENVO:grassland_soil	Soil	source	NA	n	5/28/08	Pyrosequencing-Based Assessment of Soil pH as a Predictor of Soil Bacterial Community Structure at the Continental Scale	NA	FA6P1OK	NA	mimarks-survey	NA	NA	NA	182.4	0	NA	"Kohala Peninsula, HI USA"	1500	2008	NA	0	NA	NA	NA	11.2	22.8	NA	NA	NA	HI3	1	"0.5, g"	NA	NA	targeted gene survey	13.285	NA	-155.7	y	0	CCME	NA	NA	NA	"Soils harbor enormously diverse bacterial populations, and soil bacterial communities can vary greatly in composition across space. However, our understanding of the specific changes in soil bacterial community structure that occur across larger spatial scales is limited because most previous work has focused on either surveying a relatively small number of soils in detail or analyzing a larger number of soils with techniques that provide little detail about the phylogenetic structure of the bacterial communities. Here we used a bar-coded pyrosequencing technique to characterize bacterial communities in 88 soils from across North and South America, obtaining an average of 1,501 sequences per soil. We found that overall bacterial community composition, as measured by pairwise UniFrac distances, was significantly correlated with differences in soil pH (r = 0.79), largely driven by changes in the relative abundances of Acidobacteria, Actinobacteria, and Bacteroidetes across the range of soil pHs. In addition, soil pH explains a significant portion of the variability associated with observed changes in the phylogenetic structure within each dominant lineage. The overall phylogenetic diversity of the bacterial communities was also correlated with soil pH (R2 = 0.50), with peak diversity in soils with near-neutral pHs. Together, these results suggest that the structure of soil bacterial communities is predictable, to some degree, across larger spatial scales, and the effect of soil pH on bacterial community composition is evident at even relatively coarse levels of taxonomic resolution."	ENVO:grassland soil	TCAG	Pyrosequencing_based_assessment_of_soil_pH_as_a_predictor_of_soil_bacterial_community_structure_at_the_continental_scale	HI3_V2	0	bacterial biogeography	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19502440	NA	NA	NA	Chris Lauber	CCME	GAZ:United States of America	loamy sand	lauber_88_soils	0-0.05	NA	NA	25	NA	n	NA	NA	Lauber_88_soils	pyrosequencing	NA	NA	108	soil metagenome	NA	"ENVO:Tropical and subtropical grasslands, savannas, and shrubland biome"	FLX	HI3_V2	NA	1000	6.53	NA	Lauber88Soils	Pyrosequencing_based_assessment_of_soil_pH_as_a_predictor_of_soil_bacterial_community_structure_at_the_continental_scale	410658	CCME	andisol	NA	NA	103	19502440	lauber_88_soils	NA	y	NA	HI3_V2	ENVO:soil	16S rRNA	NA	NA	V2	CCME	V2	NA	NA	Engencore	NA	20.08333333	NA	NA	soil metagenome
HI4.141735	ACGCTCATGGAT	CATGCTGCCTCCCGTAGGAGT	chris.lauber@gmail.com	NA	NA	ENVO:grassland_soil	Soil	source	NA	n	5/28/08	Pyrosequencing-Based Assessment of Soil pH as a Predictor of Soil Bacterial Community Structure at the Continental Scale	NA	FA6P1OK	NA	mimarks-survey	NA	NA	NA	108.2	0	NA	"Kohala Peninsula, HI USA"	1500	2008	NA	0	NA	NA	NA	15.1	22.8	NA	NA	NA	HI4	1	"0.5, g"	NA	NA	targeted gene survey	22.822	NA	-155.7	y	0	CCME	NA	NA	NA	"Soils harbor enormously diverse bacterial populations, and soil bacterial communities can vary greatly in composition across space. However, our understanding of the specific changes in soil bacterial community structure that occur across larger spatial scales is limited because most previous work has focused on either surveying a relatively small number of soils in detail or analyzing a larger number of soils with techniques that provide little detail about the phylogenetic structure of the bacterial communities. Here we used a bar-coded pyrosequencing technique to characterize bacterial communities in 88 soils from across North and South America, obtaining an average of 1,501 sequences per soil. We found that overall bacterial community composition, as measured by pairwise UniFrac distances, was significantly correlated with differences in soil pH (r = 0.79), largely driven by changes in the relative abundances of Acidobacteria, Actinobacteria, and Bacteroidetes across the range of soil pHs. In addition, soil pH explains a significant portion of the variability associated with observed changes in the phylogenetic structure within each dominant lineage. The overall phylogenetic diversity of the bacterial communities was also correlated with soil pH (R2 = 0.50), with peak diversity in soils with near-neutral pHs. Together, these results suggest that the structure of soil bacterial communities is predictable, to some degree, across larger spatial scales, and the effect of soil pH on bacterial community composition is evident at even relatively coarse levels of taxonomic resolution."	ENVO:grassland soil	TCAG	Pyrosequencing_based_assessment_of_soil_pH_as_a_predictor_of_soil_bacterial_community_structure_at_the_continental_scale	HI4_V2	0	bacterial biogeography	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19502440	NA	NA	NA	Chris Lauber	CCME	GAZ:United States of America	loam	lauber_88_soils	0-0.05	NA	NA	52	NA	n	NA	NA	Lauber_88_soils	pyrosequencing	NA	NA	-392	soil metagenome	NA	"ENVO:Tropical and subtropical grasslands, savannas, and shrubland biome"	FLX	HI4_V2	NA	1500	4.92	NA	Lauber88Soils	Pyrosequencing_based_assessment_of_soil_pH_as_a_predictor_of_soil_bacterial_community_structure_at_the_continental_scale	410658	CCME	andisol	NA	NA	103	19502440	lauber_88_soils	NA	y	NA	HI4_V2	ENVO:soil	16S rRNA	NA	NA	V2	CCME	V2	NA	NA	Engencore	NA	20.08333333	NA	NA	soil metagenome
IE1.141648	ACGTACTCAGTG	CATGCTGCCTCCCGTAGGAGT	chris.lauber@gmail.com	NA	NA	ENVO:grassland_soil	Soil	source	NA	n	5/28/08	Pyrosequencing-Based Assessment of Soil pH as a Predictor of Soil Bacterial Community Structure at the Continental Scale	NA	FA6P1OK	NA	mimarks-survey	NA	NA	NA	27	0	NA	"Institute for Ecosystem Studies, NY USA"	75	2008	NA	0	NA	NA	NA	11.8	8.6	NA	NA	NA	IE1	1	"0.5, g"	NA	NA	targeted gene survey	7.536	NA	-73.75	y	0	CCME	NA	NA	NA	"Soils harbor enormously diverse bacterial populations, and soil bacterial communities can vary greatly in composition across space. However, our understanding of the specific changes in soil bacterial community structure that occur across larger spatial scales is limited because most previous work has focused on either surveying a relatively small number of soils in detail or analyzing a larger number of soils with techniques that provide little detail about the phylogenetic structure of the bacterial communities. Here we used a bar-coded pyrosequencing technique to characterize bacterial communities in 88 soils from across North and South America, obtaining an average of 1,501 sequences per soil. We found that overall bacterial community composition, as measured by pairwise UniFrac distances, was significantly correlated with differences in soil pH (r = 0.79), largely driven by changes in the relative abundances of Acidobacteria, Actinobacteria, and Bacteroidetes across the range of soil pHs. In addition, soil pH explains a significant portion of the variability associated with observed changes in the phylogenetic structure within each dominant lineage. The overall phylogenetic diversity of the bacterial communities was also correlated with soil pH (R2 = 0.50), with peak diversity in soils with near-neutral pHs. Together, these results suggest that the structure of soil bacterial communities is predictable, to some degree, across larger spatial scales, and the effect of soil pH on bacterial community composition is evident at even relatively coarse levels of taxonomic resolution."	ENVO:grassland soil	TCAG	Pyrosequencing_based_assessment_of_soil_pH_as_a_predictor_of_soil_bacterial_community_structure_at_the_continental_scale	IE1_V2	0	bacterial biogeography	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19502440	NA	NA	NA	Chris Lauber	CCME	GAZ:United States of America	sandy loam	lauber_88_soils	0-0.05	NA	NA	49	NA	n	NA	NA	Lauber_88_soils	pyrosequencing	NA	NA	-610	soil metagenome	NA	ENVO:Temperate grasslands	FLX	IE1_V2	NA	1200	5.27	NA	Lauber88Soils	Pyrosequencing_based_assessment_of_soil_pH_as_a_predictor_of_soil_bacterial_community_structure_at_the_continental_scale	410658	CCME	inceptisol	NA	NA	103	19502440	lauber_88_soils	NA	y	NA	IE1_V2	ENVO:soil	16S rRNA	NA	NA	V2	CCME	V2	NA	NA	Engencore	NA	41.8	NA	NA	soil metagenome
IE2.141655	ACGTCTGTAGCA	CATGCTGCCTCCCGTAGGAGT	chris.lauber@gmail.com	NA	NA	ENVO:grassland_soil	Soil	source	NA	n	5/28/08	Pyrosequencing-Based Assessment of Soil pH as a Predictor of Soil Bacterial Community Structure at the Continental Scale	NA	FA6P1OK	NA	mimarks-survey	NA	NA	NA	40.7	0	NA	"Institute for Ecosystem Studies, NY USA"	75	2008	NA	0	NA	NA	NA	13.2	8.6	NA	NA	NA	IE2	1	"0.5, g"	NA	NA	targeted gene survey	11.422	NA	-73.75	y	0	CCME	NA	NA	NA	"Soils harbor enormously diverse bacterial populations, and soil bacterial communities can vary greatly in composition across space. However, our understanding of the specific changes in soil bacterial community structure that occur across larger spatial scales is limited because most previous work has focused on either surveying a relatively small number of soils in detail or analyzing a larger number of soils with techniques that provide little detail about the phylogenetic structure of the bacterial communities. Here we used a bar-coded pyrosequencing technique to characterize bacterial communities in 88 soils from across North and South America, obtaining an average of 1,501 sequences per soil. We found that overall bacterial community composition, as measured by pairwise UniFrac distances, was significantly correlated with differences in soil pH (r = 0.79), largely driven by changes in the relative abundances of Acidobacteria, Actinobacteria, and Bacteroidetes across the range of soil pHs. In addition, soil pH explains a significant portion of the variability associated with observed changes in the phylogenetic structure within each dominant lineage. The overall phylogenetic diversity of the bacterial communities was also correlated with soil pH (R2 = 0.50), with peak diversity in soils with near-neutral pHs. Together, these results suggest that the structure of soil bacterial communities is predictable, to some degree, across larger spatial scales, and the effect of soil pH on bacterial community composition is evident at even relatively coarse levels of taxonomic resolution."	ENVO:grassland soil	TCAG	Pyrosequencing_based_assessment_of_soil_pH_as_a_predictor_of_soil_bacterial_community_structure_at_the_continental_scale	IE2_V2	0	bacterial biogeography	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19502440	NA	NA	NA	Chris Lauber	CCME	GAZ:United States of America	sandy loam	lauber_88_soils	0-0.05	NA	NA	49	NA	n	NA	NA	Lauber_88_soils	pyrosequencing	NA	NA	-610	soil metagenome	NA	ENVO:Temperate grasslands	FLX	IE2_V2	NA	1200	5.52	NA	Lauber88Soils	Pyrosequencing_based_assessment_of_soil_pH_as_a_predictor_of_soil_bacterial_community_structure_at_the_continental_scale	410658	CCME	inceptisol	NA	NA	103	19502440	lauber_88_soils	NA	y	NA	IE2_V2	ENVO:soil	16S rRNA	NA	NA	V2	CCME	V2	NA	NA	Engencore	NA	41.8	NA	NA	soil metagenome
KP1.141713	ACTACAGCCTAT	CATGCTGCCTCCCGTAGGAGT	chris.lauber@gmail.com	NA	NA	ENVO:grassland_soil	Soil	source	NA	n	5/28/08	Pyrosequencing-Based Assessment of Soil pH as a Predictor of Soil Bacterial Community Structure at the Continental Scale	NA	FA6P1OK	NA	mimarks-survey	NA	NA	NA	61.2	0	NA	"Konza Prairie LTER, KS USA"	100	2008	NA	0	NA	NA	NA	13.6	12.5	NA	NA	NA	KP1	1	"0.5, g"	NA	NA	targeted gene survey	4.324	NA	-96.6	y	0	CCME	NA	NA	NA	"Soils harbor enormously diverse bacterial populations, and soil bacterial communities can vary greatly in composition across space. However, our understanding of the specific changes in soil bacterial community structure that occur across larger spatial scales is limited because most previous work has focused on either surveying a relatively small number of soils in detail or analyzing a larger number of soils with techniques that provide little detail about the phylogenetic structure of the bacterial communities. Here we used a bar-coded pyrosequencing technique to characterize bacterial communities in 88 soils from across North and South America, obtaining an average of 1,501 sequences per soil. We found that overall bacterial community composition, as measured by pairwise UniFrac distances, was significantly correlated with differences in soil pH (r = 0.79), largely driven by changes in the relative abundances of Acidobacteria, Actinobacteria, and Bacteroidetes across the range of soil pHs. In addition, soil pH explains a significant portion of the variability associated with observed changes in the phylogenetic structure within each dominant lineage. The overall phylogenetic diversity of the bacterial communities was also correlated with soil pH (R2 = 0.50), with peak diversity in soils with near-neutral pHs. Together, these results suggest that the structure of soil bacterial communities is predictable, to some degree, across larger spatial scales, and the effect of soil pH on bacterial community composition is evident at even relatively coarse levels of taxonomic resolution."	ENVO:grassland soil	TCAG	Pyrosequencing_based_assessment_of_soil_pH_as_a_predictor_of_soil_bacterial_community_structure_at_the_continental_scale	KP1_V2	0	bacterial biogeography	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19502440	NA	NA	NA	Chris Lauber	CCME	GAZ:United States of America	silt loam	lauber_88_soils	0-0.05	NA	NA	78	NA	n	NA	NA	Lauber_88_soils	pyrosequencing	NA	NA	-81	soil metagenome	NA	ENVO:Temperate grasslands	FLX	KP1_V2	NA	835	6.37	NA	Lauber88Soils	Pyrosequencing_based_assessment_of_soil_pH_as_a_predictor_of_soil_bacterial_community_structure_at_the_continental_scale	410658	CCME	mollisol	NA	NA	103	19502440	lauber_88_soils	NA	y	NA	KP1_V2	ENVO:soil	16S rRNA	NA	NA	V2	CCME	V2	NA	NA	Engencore	NA	39.1	NA	NA	soil metagenome
KP4.141733	ACTAGCTCCATA	CATGCTGCCTCCCGTAGGAGT	chris.lauber@gmail.com	NA	NA	ENVO:grassland_soil	Soil	source	NA	n	5/28/08	Pyrosequencing-Based Assessment of Soil pH as a Predictor of Soil Bacterial Community Structure at the Continental Scale	NA	FA6P1OK	NA	mimarks-survey	NA	NA	NA	39.8	0	NA	"Konza Prairie LTER, KS USA"	100	2008	NA	0	NA	NA	NA	13.892	12.5	NA	NA	NA	KP4	1	"0.5, g"	NA	NA	targeted gene survey	16	NA	-96.6	y	0	CCME	NA	NA	NA	"Soils harbor enormously diverse bacterial populations, and soil bacterial communities can vary greatly in composition across space. However, our understanding of the specific changes in soil bacterial community structure that occur across larger spatial scales is limited because most previous work has focused on either surveying a relatively small number of soils in detail or analyzing a larger number of soils with techniques that provide little detail about the phylogenetic structure of the bacterial communities. Here we used a bar-coded pyrosequencing technique to characterize bacterial communities in 88 soils from across North and South America, obtaining an average of 1,501 sequences per soil. We found that overall bacterial community composition, as measured by pairwise UniFrac distances, was significantly correlated with differences in soil pH (r = 0.79), largely driven by changes in the relative abundances of Acidobacteria, Actinobacteria, and Bacteroidetes across the range of soil pHs. In addition, soil pH explains a significant portion of the variability associated with observed changes in the phylogenetic structure within each dominant lineage. The overall phylogenetic diversity of the bacterial communities was also correlated with soil pH (R2 = 0.50), with peak diversity in soils with near-neutral pHs. Together, these results suggest that the structure of soil bacterial communities is predictable, to some degree, across larger spatial scales, and the effect of soil pH on bacterial community composition is evident at even relatively coarse levels of taxonomic resolution."	ENVO:grassland soil	TCAG	Pyrosequencing_based_assessment_of_soil_pH_as_a_predictor_of_soil_bacterial_community_structure_at_the_continental_scale	KP4_V2	0	bacterial biogeography	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19502440	NA	NA	NA	Chris Lauber	CCME	GAZ:United States of America	silt loam	lauber_88_soils	0-0.05	NA	NA	77	NA	n	NA	NA	Lauber_88_soils	pyrosequencing	NA	NA	-81	soil metagenome	NA	ENVO:shrubland	FLX	KP4_V2	NA	835	7.1	NA	Lauber88Soils	Pyrosequencing_based_assessment_of_soil_pH_as_a_predictor_of_soil_bacterial_community_structure_at_the_continental_scale	410658	CCME	mollisol	NA	NA	103	19502440	lauber_88_soils	NA	y	NA	KP4_V2	ENVO:soil	16S rRNA	NA	NA	V2	CCME	V2	NA	NA	Engencore	NA	39.1	NA	NA	soil metagenome
MP1.141661	ACTGATCCTAGT	CATGCTGCCTCCCGTAGGAGT	chris.lauber@gmail.com	NA	NA	ENVO:grassland_soil	Soil	source	NA	n	5/28/08	Pyrosequencing-Based Assessment of Soil pH as a Predictor of Soil Bacterial Community Structure at the Continental Scale	NA	FA6P1OK	NA	mimarks-survey	NA	NA	NA	107	0	NA	"Mary's Peak, OR USA"	1300	2008	NA	0	NA	NA	NA	14.7	8.8	NA	NA	NA	MP1	1	"0.5, g"	NA	NA	targeted gene survey	5.636	NA	-123.5333333	y	0	CCME	NA	NA	NA	"Soils harbor enormously diverse bacterial populations, and soil bacterial communities can vary greatly in composition across space. However, our understanding of the specific changes in soil bacterial community structure that occur across larger spatial scales is limited because most previous work has focused on either surveying a relatively small number of soils in detail or analyzing a larger number of soils with techniques that provide little detail about the phylogenetic structure of the bacterial communities. Here we used a bar-coded pyrosequencing technique to characterize bacterial communities in 88 soils from across North and South America, obtaining an average of 1,501 sequences per soil. We found that overall bacterial community composition, as measured by pairwise UniFrac distances, was significantly correlated with differences in soil pH (r = 0.79), largely driven by changes in the relative abundances of Acidobacteria, Actinobacteria, and Bacteroidetes across the range of soil pHs. In addition, soil pH explains a significant portion of the variability associated with observed changes in the phylogenetic structure within each dominant lineage. The overall phylogenetic diversity of the bacterial communities was also correlated with soil pH (R2 = 0.50), with peak diversity in soils with near-neutral pHs. Together, these results suggest that the structure of soil bacterial communities is predictable, to some degree, across larger spatial scales, and the effect of soil pH on bacterial community composition is evident at even relatively coarse levels of taxonomic resolution."	ENVO:grassland soil	TCAG	Pyrosequencing_based_assessment_of_soil_pH_as_a_predictor_of_soil_bacterial_community_structure_at_the_continental_scale	MP1_V2	0	bacterial biogeography	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19502440	NA	NA	NA	Chris Lauber	CCME	GAZ:United States of America	sandy loam	lauber_88_soils	0-0.05	NA	NA	41	NA	n	NA	NA	Lauber_88_soils	pyrosequencing	NA	NA	-1721	soil metagenome	NA	ENVO:Temperate grasslands	FLX	MP1_V2	NA	2200	4.56	NA	Lauber88Soils	Pyrosequencing_based_assessment_of_soil_pH_as_a_predictor_of_soil_bacterial_community_structure_at_the_continental_scale	410658	CCME	inceptisol	NA	NA	103	19502440	lauber_88_soils	NA	y	NA	MP1_V2	ENVO:soil	16S rRNA	NA	NA	V2	CCME	V2	NA	NA	Engencore	NA	49.46666667	NA	NA	soil metagenome
RT2.141710	AGAGTCCTGAGC	CATGCTGCCTCCCGTAGGAGT	chris.lauber@gmail.com	NA	NA	ENVO:grassland_soil	Soil	source	NA	n	5/28/08	Pyrosequencing-Based Assessment of Soil pH as a Predictor of Soil Bacterial Community Structure at the Continental Scale	NA	FA6P1OK	NA	mimarks-survey	NA	NA	NA	37.5	0	NA	"USDA Grassland Research Center, Riesel, TX USA"	50	2008	NA	0	NA	NA	NA	12	18.1	NA	NA	NA	RT2	1	"0.5, g"	NA	NA	targeted gene survey	1.646	NA	-96.86666667	y	15.8	CCME	NA	NA	NA	"Soils harbor enormously diverse bacterial populations, and soil bacterial communities can vary greatly in composition across space. However, our understanding of the specific changes in soil bacterial community structure that occur across larger spatial scales is limited because most previous work has focused on either surveying a relatively small number of soils in detail or analyzing a larger number of soils with techniques that provide little detail about the phylogenetic structure of the bacterial communities. Here we used a bar-coded pyrosequencing technique to characterize bacterial communities in 88 soils from across North and South America, obtaining an average of 1,501 sequences per soil. We found that overall bacterial community composition, as measured by pairwise UniFrac distances, was significantly correlated with differences in soil pH (r = 0.79), largely driven by changes in the relative abundances of Acidobacteria, Actinobacteria, and Bacteroidetes across the range of soil pHs. In addition, soil pH explains a significant portion of the variability associated with observed changes in the phylogenetic structure within each dominant lineage. The overall phylogenetic diversity of the bacterial communities was also correlated with soil pH (R2 = 0.50), with peak diversity in soils with near-neutral pHs. Together, these results suggest that the structure of soil bacterial communities is predictable, to some degree, across larger spatial scales, and the effect of soil pH on bacterial community composition is evident at even relatively coarse levels of taxonomic resolution."	ENVO:grassland soil	TCAG	Pyrosequencing_based_assessment_of_soil_pH_as_a_predictor_of_soil_bacterial_community_structure_at_the_continental_scale	RT2_V2	0	bacterial biogeography	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19502440	NA	NA	NA	Chris Lauber	CCME	GAZ:United States of America	silty clay loam	lauber_88_soils	0-0.05	NA	NA	80	NA	n	NA	NA	Lauber_88_soils	pyrosequencing	NA	NA	135	soil metagenome	NA	ENVO:Temperate grasslands	FLX	RT2_V2	NA	840	8.07	NA	Lauber88Soils	Pyrosequencing_based_assessment_of_soil_pH_as_a_predictor_of_soil_bacterial_community_structure_at_the_continental_scale	410658	CCME	mollisol	NA	NA	103	19502440	lauber_88_soils	NA	y	NA	RT2_V2	ENVO:soil	16S rRNA	NA	NA	V2	CCME	V2	NA	NA	Engencore	NA	31.46666667	NA	NA	soil metagenome
SP2.141678	AGCGCTGATGTG	CATGCTGCCTCCCGTAGGAGT	chris.lauber@gmail.com	NA	NA	ENVO:grassland_soil	Soil	source	NA	n	5/28/08	Pyrosequencing-Based Assessment of Soil pH as a Predictor of Soil Bacterial Community Structure at the Continental Scale	NA	FA6P1OK	NA	mimarks-survey	NA	NA	NA	81	0	NA	"Sequoia National Park, CA USA"	3215	2008	NA	0	NA	NA	NA	15.9	3.6	NA	NA	NA	SP2	1	"0.5, g"	NA	NA	targeted gene survey	6.535	NA	-118.6333333	y	0	CCME	NA	NA	NA	"Soils harbor enormously diverse bacterial populations, and soil bacterial communities can vary greatly in composition across space. However, our understanding of the specific changes in soil bacterial community structure that occur across larger spatial scales is limited because most previous work has focused on either surveying a relatively small number of soils in detail or analyzing a larger number of soils with techniques that provide little detail about the phylogenetic structure of the bacterial communities. Here we used a bar-coded pyrosequencing technique to characterize bacterial communities in 88 soils from across North and South America, obtaining an average of 1,501 sequences per soil. We found that overall bacterial community composition, as measured by pairwise UniFrac distances, was significantly correlated with differences in soil pH (r = 0.79), largely driven by changes in the relative abundances of Acidobacteria, Actinobacteria, and Bacteroidetes across the range of soil pHs. In addition, soil pH explains a significant portion of the variability associated with observed changes in the phylogenetic structure within each dominant lineage. The overall phylogenetic diversity of the bacterial communities was also correlated with soil pH (R2 = 0.50), with peak diversity in soils with near-neutral pHs. Together, these results suggest that the structure of soil bacterial communities is predictable, to some degree, across larger spatial scales, and the effect of soil pH on bacterial community composition is evident at even relatively coarse levels of taxonomic resolution."	ENVO:grassland soil	TCAG	Pyrosequencing_based_assessment_of_soil_pH_as_a_predictor_of_soil_bacterial_community_structure_at_the_continental_scale	SP2_V2	0	bacterial biogeography	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19502440	NA	NA	NA	Chris Lauber	CCME	GAZ:United States of America	loamy sand	lauber_88_soils	0-0.05	NA	NA	24	NA	n	NA	NA	Lauber_88_soils	pyrosequencing	NA	NA	-402	soil metagenome	NA	ENVO:Temperate grasslands	FLX	SP2_V2	NA	750	5.13	NA	Lauber88Soils	Pyrosequencing_based_assessment_of_soil_pH_as_a_predictor_of_soil_bacterial_community_structure_at_the_continental_scale	410658	CCME	entisol	NA	NA	103	19502440	lauber_88_soils	NA	y	NA	SP2_V2	ENVO:soil	16S rRNA	NA	NA	V2	CCME	V2	NA	NA	Engencore	NA	36.61666667	NA	NA	soil metagenome
TL1.141653	AGCTTGACAGCT	CATGCTGCCTCCCGTAGGAGT	chris.lauber@gmail.com	NA	NA	ENVO:grassland_soil	Soil	source	NA	n	5/28/08	Pyrosequencing-Based Assessment of Soil pH as a Predictor of Soil Bacterial Community Structure at the Continental Scale	NA	FA6P1OK	NA	mimarks-survey	NA	NA	NA	70.2	0	NA	"Toolik Lake LTER, AK USA"	894	2008	NA	0	NA	NA	NA	18.7	-9.3	NA	NA	NA	TL1	1	"0.5, g"	NA	NA	targeted gene survey	9.241	NA	-149.5833333	y	0	CCME	NA	NA	NA	"Soils harbor enormously diverse bacterial populations, and soil bacterial communities can vary greatly in composition across space. However, our understanding of the specific changes in soil bacterial community structure that occur across larger spatial scales is limited because most previous work has focused on either surveying a relatively small number of soils in detail or analyzing a larger number of soils with techniques that provide little detail about the phylogenetic structure of the bacterial communities. Here we used a bar-coded pyrosequencing technique to characterize bacterial communities in 88 soils from across North and South America, obtaining an average of 1,501 sequences per soil. We found that overall bacterial community composition, as measured by pairwise UniFrac distances, was significantly correlated with differences in soil pH (r = 0.79), largely driven by changes in the relative abundances of Acidobacteria, Actinobacteria, and Bacteroidetes across the range of soil pHs. In addition, soil pH explains a significant portion of the variability associated with observed changes in the phylogenetic structure within each dominant lineage. The overall phylogenetic diversity of the bacterial communities was also correlated with soil pH (R2 = 0.50), with peak diversity in soils with near-neutral pHs. Together, these results suggest that the structure of soil bacterial communities is predictable, to some degree, across larger spatial scales, and the effect of soil pH on bacterial community composition is evident at even relatively coarse levels of taxonomic resolution."	ENVO:grassland soil	TCAG	Pyrosequencing_based_assessment_of_soil_pH_as_a_predictor_of_soil_bacterial_community_structure_at_the_continental_scale	TL1_V2	0	bacterial biogeography	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19502440	NA	NA	NA	Chris Lauber	CCME	GAZ:United States of America	loam	lauber_88_soils	0-0.05	NA	NA	57	NA	n	NA	NA	Lauber_88_soils	pyrosequencing	NA	NA	-212	soil metagenome	NA	ENVO:grassland	FLX	TL1_V2	NA	400	4.58	NA	Lauber88Soils	Pyrosequencing_based_assessment_of_soil_pH_as_a_predictor_of_soil_bacterial_community_structure_at_the_continental_scale	410658	CCME	gelisol	NA	NA	103	19502440	lauber_88_soils	NA	y	NA	TL1_V2	ENVO:soil	16S rRNA	NA	NA	V2	CCME	V2	NA	NA	Engencore	NA	68.63333333	NA	NA	soil metagenome
VC2.141694	AGTACGCTCGAG	CATGCTGCCTCCCGTAGGAGT	chris.lauber@gmail.com	NA	NA	ENVO:grassland_soil	Soil	source	NA	n	5/28/08	Pyrosequencing-Based Assessment of Soil pH as a Predictor of Soil Bacterial Community Structure at the Continental Scale	NA	FA6P1OK	NA	mimarks-survey	NA	NA	NA	34.4	0	NA	"Valles Caldera, NM USA"	2733	2008	NA	0	NA	NA	NA	11.5	2.5	NA	NA	NA	VC2	1	"0.5, g"	NA	NA	targeted gene survey	6.265	NA	-106.55	y	0	CCME	NA	NA	NA	"Soils harbor enormously diverse bacterial populations, and soil bacterial communities can vary greatly in composition across space. However, our understanding of the specific changes in soil bacterial community structure that occur across larger spatial scales is limited because most previous work has focused on either surveying a relatively small number of soils in detail or analyzing a larger number of soils with techniques that provide little detail about the phylogenetic structure of the bacterial communities. Here we used a bar-coded pyrosequencing technique to characterize bacterial communities in 88 soils from across North and South America, obtaining an average of 1,501 sequences per soil. We found that overall bacterial community composition, as measured by pairwise UniFrac distances, was significantly correlated with differences in soil pH (r = 0.79), largely driven by changes in the relative abundances of Acidobacteria, Actinobacteria, and Bacteroidetes across the range of soil pHs. In addition, soil pH explains a significant portion of the variability associated with observed changes in the phylogenetic structure within each dominant lineage. The overall phylogenetic diversity of the bacterial communities was also correlated with soil pH (R2 = 0.50), with peak diversity in soils with near-neutral pHs. Together, these results suggest that the structure of soil bacterial communities is predictable, to some degree, across larger spatial scales, and the effect of soil pH on bacterial community composition is evident at even relatively coarse levels of taxonomic resolution."	ENVO:grassland soil	TCAG	Pyrosequencing_based_assessment_of_soil_pH_as_a_predictor_of_soil_bacterial_community_structure_at_the_continental_scale	VC2_V2	0	bacterial biogeography	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19502440	NA	NA	NA	Chris Lauber	CCME	GAZ:United States of America	loam	lauber_88_soils	0-0.05	NA	NA	59	NA	n	NA	NA	Lauber_88_soils	pyrosequencing	NA	NA	-175	soil metagenome	NA	ENVO:grassland	FLX	VC2_V2	NA	500	5.99	NA	Lauber88Soils	Pyrosequencing_based_assessment_of_soil_pH_as_a_predictor_of_soil_bacterial_community_structure_at_the_continental_scale	410658	CCME	mollisol	NA	NA	103	19502440	lauber_88_soils	NA	y	NA	VC2_V2	ENVO:soil	16S rRNA	NA	NA	V2	CCME	V2	NA	NA	Engencore	NA	35.9	NA	NA	soil metagenome
JT1.141699	ACGTTAGCACAC	CATGCTGCCTCCCGTAGGAGT	chris.lauber@gmail.com	NA	NA	ENVO:shrubland	Soil	source	NA	n	5/28/08	Pyrosequencing-Based Assessment of Soil pH as a Predictor of Soil Bacterial Community Structure at the Continental Scale	NA	FA6P1OK	NA	mimarks-survey	NA	NA	NA	6.9	0	NA	"Joshua Tree National Park, CA USA"	1360	2008	NA	0	NA	NA	NA	11.668	16	NA	NA	NA	JT1	1	"0.5, g"	NA	NA	targeted gene survey	11.7	NA	-116.0666667	y	0	CCME	NA	NA	NA	"Soils harbor enormously diverse bacterial populations, and soil bacterial communities can vary greatly in composition across space. However, our understanding of the specific changes in soil bacterial community structure that occur across larger spatial scales is limited because most previous work has focused on either surveying a relatively small number of soils in detail or analyzing a larger number of soils with techniques that provide little detail about the phylogenetic structure of the bacterial communities. Here we used a bar-coded pyrosequencing technique to characterize bacterial communities in 88 soils from across North and South America, obtaining an average of 1,501 sequences per soil. We found that overall bacterial community composition, as measured by pairwise UniFrac distances, was significantly correlated with differences in soil pH (r = 0.79), largely driven by changes in the relative abundances of Acidobacteria, Actinobacteria, and Bacteroidetes across the range of soil pHs. In addition, soil pH explains a significant portion of the variability associated with observed changes in the phylogenetic structure within each dominant lineage. The overall phylogenetic diversity of the bacterial communities was also correlated with soil pH (R2 = 0.50), with peak diversity in soils with near-neutral pHs. Together, these results suggest that the structure of soil bacterial communities is predictable, to some degree, across larger spatial scales, and the effect of soil pH on bacterial community composition is evident at even relatively coarse levels of taxonomic resolution."	ENVO:shrubland	TCAG	Pyrosequencing_based_assessment_of_soil_pH_as_a_predictor_of_soil_bacterial_community_structure_at_the_continental_scale	JT1_V2	0	bacterial biogeography	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19502440	NA	NA	NA	Chris Lauber	CCME	GAZ:United States of America	loamy sand	lauber_88_soils	0-0.05	NA	NA	18	NA	n	NA	NA	Lauber_88_soils	pyrosequencing	NA	NA	1032	soil metagenome	NA	ENVO:shrubland	FLX	JT1_V2	NA	90	7.6	NA	Lauber88Soils	Pyrosequencing_based_assessment_of_soil_pH_as_a_predictor_of_soil_bacterial_community_structure_at_the_continental_scale	410658	CCME	aridisol	NA	NA	103	19502440	lauber_88_soils	NA	y	NA	JT1_V2	ENVO:soil	16S rRNA	NA	NA	V2	CCME	V2	NA	NA	Engencore	NA	33.96666667	NA	NA	soil metagenome
KP3.141658	ACTACGTGTGGT	CATGCTGCCTCCCGTAGGAGT	chris.lauber@gmail.com	NA	NA	ENVO:shrubland	Soil	source	NA	n	5/28/08	Pyrosequencing-Based Assessment of Soil pH as a Predictor of Soil Bacterial Community Structure at the Continental Scale	NA	FA6P1OK	NA	mimarks-survey	NA	NA	NA	68.9	0	NA	"Konza Prairie LTER, KS USA"	100	2008	NA	0	NA	NA	NA	17.7	12.5	NA	NA	NA	KP3	1	"0.5, g"	NA	NA	targeted gene survey	1.948	NA	-96.6	y	0	CCME	NA	NA	NA	"Soils harbor enormously diverse bacterial populations, and soil bacterial communities can vary greatly in composition across space. However, our understanding of the specific changes in soil bacterial community structure that occur across larger spatial scales is limited because most previous work has focused on either surveying a relatively small number of soils in detail or analyzing a larger number of soils with techniques that provide little detail about the phylogenetic structure of the bacterial communities. Here we used a bar-coded pyrosequencing technique to characterize bacterial communities in 88 soils from across North and South America, obtaining an average of 1,501 sequences per soil. We found that overall bacterial community composition, as measured by pairwise UniFrac distances, was significantly correlated with differences in soil pH (r = 0.79), largely driven by changes in the relative abundances of Acidobacteria, Actinobacteria, and Bacteroidetes across the range of soil pHs. In addition, soil pH explains a significant portion of the variability associated with observed changes in the phylogenetic structure within each dominant lineage. The overall phylogenetic diversity of the bacterial communities was also correlated with soil pH (R2 = 0.50), with peak diversity in soils with near-neutral pHs. Together, these results suggest that the structure of soil bacterial communities is predictable, to some degree, across larger spatial scales, and the effect of soil pH on bacterial community composition is evident at even relatively coarse levels of taxonomic resolution."	ENVO:shrubland	TCAG	Pyrosequencing_based_assessment_of_soil_pH_as_a_predictor_of_soil_bacterial_community_structure_at_the_continental_scale	KP3_V2	0	bacterial biogeography	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19502440	NA	NA	NA	Chris Lauber	CCME	GAZ:United States of America	silt loam	lauber_88_soils	0-0.05	NA	NA	76	NA	n	NA	NA	Lauber_88_soils	pyrosequencing	NA	NA	-81	soil metagenome	NA	ENVO:shrubland	FLX	KP3_V2	NA	835	7.92	NA	Lauber88Soils	Pyrosequencing_based_assessment_of_soil_pH_as_a_predictor_of_soil_bacterial_community_structure_at_the_continental_scale	410658	CCME	mollisol	NA	NA	103	19502440	lauber_88_soils	NA	y	NA	KP3_V2	ENVO:soil	16S rRNA	NA	NA	V2	CCME	V2	NA	NA	Engencore	NA	39.1	NA	NA	soil metagenome
MD2.141689	ACTCGATTCGAT	CATGCTGCCTCCCGTAGGAGT	chris.lauber@gmail.com	NA	NA	ENVO:shrubland	Soil	source	NA	n	5/28/08	Pyrosequencing-Based Assessment of Soil pH as a Predictor of Soil Bacterial Community Structure at the Continental Scale	NA	FA6P1OK	NA	mimarks-survey	NA	NA	NA	4.2	0	NA	"Mojave Desert, CA USA"	967	2008	NA	0	NA	NA	NA	8.8	21	NA	NA	NA	MD2	1	"0.5, g"	NA	NA	targeted gene survey	1.97	NA	-115.6	y	0	CCME	NA	NA	NA	"Soils harbor enormously diverse bacterial populations, and soil bacterial communities can vary greatly in composition across space. However, our understanding of the specific changes in soil bacterial community structure that occur across larger spatial scales is limited because most previous work has focused on either surveying a relatively small number of soils in detail or analyzing a larger number of soils with techniques that provide little detail about the phylogenetic structure of the bacterial communities. Here we used a bar-coded pyrosequencing technique to characterize bacterial communities in 88 soils from across North and South America, obtaining an average of 1,501 sequences per soil. We found that overall bacterial community composition, as measured by pairwise UniFrac distances, was significantly correlated with differences in soil pH (r = 0.79), largely driven by changes in the relative abundances of Acidobacteria, Actinobacteria, and Bacteroidetes across the range of soil pHs. In addition, soil pH explains a significant portion of the variability associated with observed changes in the phylogenetic structure within each dominant lineage. The overall phylogenetic diversity of the bacterial communities was also correlated with soil pH (R2 = 0.50), with peak diversity in soils with near-neutral pHs. Together, these results suggest that the structure of soil bacterial communities is predictable, to some degree, across larger spatial scales, and the effect of soil pH on bacterial community composition is evident at even relatively coarse levels of taxonomic resolution."	ENVO:shrubland	TCAG	Pyrosequencing_based_assessment_of_soil_pH_as_a_predictor_of_soil_bacterial_community_structure_at_the_continental_scale	MD2_V2	0	bacterial biogeography	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19502440	NA	NA	NA	Chris Lauber	CCME	GAZ:United States of America	sandy loam	lauber_88_soils	0-0.05	NA	NA	26	NA	n	NA	NA	Lauber_88_soils	pyrosequencing	NA	NA	1054	soil metagenome	NA	ENVO:shrubland	FLX	MD2_V2	NA	150	7.65	NA	Lauber88Soils	Pyrosequencing_based_assessment_of_soil_pH_as_a_predictor_of_soil_bacterial_community_structure_at_the_continental_scale	410658	CCME	aridisol	NA	NA	103	19502440	lauber_88_soils	NA	y	NA	MD2_V2	ENVO:soil	16S rRNA	NA	NA	V2	CCME	V2	NA	NA	Engencore	NA	34.9	NA	NA	soil metagenome
MD5.141688	ACTGACAGCCAT	CATGCTGCCTCCCGTAGGAGT	chris.lauber@gmail.com	NA	NA	ENVO:shrubland	Soil	source	NA	n	5/28/08	Pyrosequencing-Based Assessment of Soil pH as a Predictor of Soil Bacterial Community Structure at the Continental Scale	NA	FA6P1OK	NA	mimarks-survey	NA	NA	NA	5.7	0	NA	"Mojave Desert, CA USA"	776	2008	NA	0	NA	NA	NA	9.5	21	NA	NA	NA	MD5	1	"0.5, g"	NA	NA	targeted gene survey	5.931	NA	-115.8666667	y	1.8	CCME	NA	NA	NA	"Soils harbor enormously diverse bacterial populations, and soil bacterial communities can vary greatly in composition across space. However, our understanding of the specific changes in soil bacterial community structure that occur across larger spatial scales is limited because most previous work has focused on either surveying a relatively small number of soils in detail or analyzing a larger number of soils with techniques that provide little detail about the phylogenetic structure of the bacterial communities. Here we used a bar-coded pyrosequencing technique to characterize bacterial communities in 88 soils from across North and South America, obtaining an average of 1,501 sequences per soil. We found that overall bacterial community composition, as measured by pairwise UniFrac distances, was significantly correlated with differences in soil pH (r = 0.79), largely driven by changes in the relative abundances of Acidobacteria, Actinobacteria, and Bacteroidetes across the range of soil pHs. In addition, soil pH explains a significant portion of the variability associated with observed changes in the phylogenetic structure within each dominant lineage. The overall phylogenetic diversity of the bacterial communities was also correlated with soil pH (R2 = 0.50), with peak diversity in soils with near-neutral pHs. Together, these results suggest that the structure of soil bacterial communities is predictable, to some degree, across larger spatial scales, and the effect of soil pH on bacterial community composition is evident at even relatively coarse levels of taxonomic resolution."	ENVO:shrubland	TCAG	Pyrosequencing_based_assessment_of_soil_pH_as_a_predictor_of_soil_bacterial_community_structure_at_the_continental_scale	MD5_V2	0	bacterial biogeography	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19502440	NA	NA	NA	Chris Lauber	CCME	GAZ:United States of America	loamy sand	lauber_88_soils	0-0.05	NA	NA	20	NA	n	NA	NA	Lauber_88_soils	pyrosequencing	NA	NA	1087	soil metagenome	NA	ENVO:shrubland	FLX	MD5_V2	NA	150	8.07	NA	Lauber88Soils	Pyrosequencing_based_assessment_of_soil_pH_as_a_predictor_of_soil_bacterial_community_structure_at_the_continental_scale	410658	CCME	aridisol	NA	NA	103	19502440	lauber_88_soils	NA	y	NA	MD5_V2	ENVO:soil	16S rRNA	NA	NA	V2	CCME	V2	NA	NA	Engencore	NA	35.2	NA	NA	soil metagenome
RT1.141654	AGAGTAGCTAAG	CATGCTGCCTCCCGTAGGAGT	chris.lauber@gmail.com	NA	NA	ENVO:shrubland	Soil	source	NA	n	5/28/08	Pyrosequencing-Based Assessment of Soil pH as a Predictor of Soil Bacterial Community Structure at the Continental Scale	NA	FA6P1OK	NA	mimarks-survey	NA	NA	NA	39.4	0	NA	"USDA Grassland Research Center, Riesel, TX USA"	50	2008	NA	0	NA	NA	NA	12.3	18.1	NA	NA	NA	RT1	1	"0.5, g"	NA	NA	targeted gene survey	1.852	NA	-96.86666667	y	18	CCME	NA	NA	NA	"Soils harbor enormously diverse bacterial populations, and soil bacterial communities can vary greatly in composition across space. However, our understanding of the specific changes in soil bacterial community structure that occur across larger spatial scales is limited because most previous work has focused on either surveying a relatively small number of soils in detail or analyzing a larger number of soils with techniques that provide little detail about the phylogenetic structure of the bacterial communities. Here we used a bar-coded pyrosequencing technique to characterize bacterial communities in 88 soils from across North and South America, obtaining an average of 1,501 sequences per soil. We found that overall bacterial community composition, as measured by pairwise UniFrac distances, was significantly correlated with differences in soil pH (r = 0.79), largely driven by changes in the relative abundances of Acidobacteria, Actinobacteria, and Bacteroidetes across the range of soil pHs. In addition, soil pH explains a significant portion of the variability associated with observed changes in the phylogenetic structure within each dominant lineage. The overall phylogenetic diversity of the bacterial communities was also correlated with soil pH (R2 = 0.50), with peak diversity in soils with near-neutral pHs. Together, these results suggest that the structure of soil bacterial communities is predictable, to some degree, across larger spatial scales, and the effect of soil pH on bacterial community composition is evident at even relatively coarse levels of taxonomic resolution."	ENVO:shrubland	TCAG	Pyrosequencing_based_assessment_of_soil_pH_as_a_predictor_of_soil_bacterial_community_structure_at_the_continental_scale	RT1_V2	0	bacterial biogeography	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19502440	NA	NA	NA	Chris Lauber	CCME	GAZ:United States of America	silty clay loam	lauber_88_soils	0-0.05	NA	NA	82	NA	n	NA	NA	Lauber_88_soils	pyrosequencing	NA	NA	135	soil metagenome	NA	ENVO:shrubland	FLX	RT1_V2	NA	840	7.92	NA	Lauber88Soils	Pyrosequencing_based_assessment_of_soil_pH_as_a_predictor_of_soil_bacterial_community_structure_at_the_continental_scale	410658	CCME	mollisol	NA	NA	103	19502440	lauber_88_soils	NA	y	NA	RT1_V2	ENVO:soil	16S rRNA	NA	NA	V2	CCME	V2	NA	NA	Engencore	NA	31.46666667	NA	NA	soil metagenome
SB1.141730	AGATCTCTGCAT	CATGCTGCCTCCCGTAGGAGT	chris.lauber@gmail.com	NA	NA	ENVO:shrubland	Soil	source	NA	n	5/28/08	Pyrosequencing-Based Assessment of Soil pH as a Predictor of Soil Bacterial Community Structure at the Continental Scale	NA	FA6P1OK	NA	mimarks-survey	NA	NA	NA	26.5	0	NA	"Santa Barbara, CA USA"	500	2008	NA	0	NA	NA	NA	16.9	15	NA	NA	NA	SB1	1	"0.5, g"	NA	NA	targeted gene survey	7.166	NA	-119.8	y	0	CCME	NA	NA	NA	"Soils harbor enormously diverse bacterial populations, and soil bacterial communities can vary greatly in composition across space. However, our understanding of the specific changes in soil bacterial community structure that occur across larger spatial scales is limited because most previous work has focused on either surveying a relatively small number of soils in detail or analyzing a larger number of soils with techniques that provide little detail about the phylogenetic structure of the bacterial communities. Here we used a bar-coded pyrosequencing technique to characterize bacterial communities in 88 soils from across North and South America, obtaining an average of 1,501 sequences per soil. We found that overall bacterial community composition, as measured by pairwise UniFrac distances, was significantly correlated with differences in soil pH (r = 0.79), largely driven by changes in the relative abundances of Acidobacteria, Actinobacteria, and Bacteroidetes across the range of soil pHs. In addition, soil pH explains a significant portion of the variability associated with observed changes in the phylogenetic structure within each dominant lineage. The overall phylogenetic diversity of the bacterial communities was also correlated with soil pH (R2 = 0.50), with peak diversity in soils with near-neutral pHs. Together, these results suggest that the structure of soil bacterial communities is predictable, to some degree, across larger spatial scales, and the effect of soil pH on bacterial community composition is evident at even relatively coarse levels of taxonomic resolution."	ENVO:shrubland	TCAG	Pyrosequencing_based_assessment_of_soil_pH_as_a_predictor_of_soil_bacterial_community_structure_at_the_continental_scale	SB1_V2	0	bacterial biogeography	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19502440	NA	NA	NA	Chris Lauber	CCME	GAZ:United States of America	loam	lauber_88_soils	0-0.05	NA	NA	54	NA	n	NA	NA	Lauber_88_soils	pyrosequencing	NA	NA	153	soil metagenome	NA	ENVO:shrubland	FLX	SB1_V2	NA	550	7.92	NA	Lauber88Soils	Pyrosequencing_based_assessment_of_soil_pH_as_a_predictor_of_soil_bacterial_community_structure_at_the_continental_scale	410658	CCME	inceptisol	NA	NA	103	19502440	lauber_88_soils	NA	y	NA	SB1_V2	ENVO:soil	16S rRNA	NA	NA	V2	CCME	V2	NA	NA	Engencore	NA	34.46666667	NA	NA	soil metagenome
SF1.141728	AGATGTTCTGCT	CATGCTGCCTCCCGTAGGAGT	chris.lauber@gmail.com	NA	NA	ENVO:shrubland	Soil	source	NA	n	5/28/08	Pyrosequencing-Based Assessment of Soil pH as a Predictor of Soil Bacterial Community Structure at the Continental Scale	NA	FA6P1OK	NA	mimarks-survey	NA	NA	NA	14.2	0	NA	"Santa Fe, NM USA"	1500	2008	NA	0	NA	NA	NA	9.92	13	NA	NA	NA	SF1	1	"0.5, g"	NA	NA	targeted gene survey	7.6	NA	-105.9333333	y	6.9	CCME	NA	NA	NA	"Soils harbor enormously diverse bacterial populations, and soil bacterial communities can vary greatly in composition across space. However, our understanding of the specific changes in soil bacterial community structure that occur across larger spatial scales is limited because most previous work has focused on either surveying a relatively small number of soils in detail or analyzing a larger number of soils with techniques that provide little detail about the phylogenetic structure of the bacterial communities. Here we used a bar-coded pyrosequencing technique to characterize bacterial communities in 88 soils from across North and South America, obtaining an average of 1,501 sequences per soil. We found that overall bacterial community composition, as measured by pairwise UniFrac distances, was significantly correlated with differences in soil pH (r = 0.79), largely driven by changes in the relative abundances of Acidobacteria, Actinobacteria, and Bacteroidetes across the range of soil pHs. In addition, soil pH explains a significant portion of the variability associated with observed changes in the phylogenetic structure within each dominant lineage. The overall phylogenetic diversity of the bacterial communities was also correlated with soil pH (R2 = 0.50), with peak diversity in soils with near-neutral pHs. Together, these results suggest that the structure of soil bacterial communities is predictable, to some degree, across larger spatial scales, and the effect of soil pH on bacterial community composition is evident at even relatively coarse levels of taxonomic resolution."	ENVO:shrubland	TCAG	Pyrosequencing_based_assessment_of_soil_pH_as_a_predictor_of_soil_bacterial_community_structure_at_the_continental_scale	SF1_V2	0	bacterial biogeography	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19502440	NA	NA	NA	Chris Lauber	CCME	GAZ:United States of America	clay loam	lauber_88_soils	0-0.05	NA	NA	60	NA	n	NA	NA	Lauber_88_soils	pyrosequencing	NA	NA	494	soil metagenome	NA	ENVO:shrubland	FLX	SF1_V2	NA	250	7.71	NA	Lauber88Soils	Pyrosequencing_based_assessment_of_soil_pH_as_a_predictor_of_soil_bacterial_community_structure_at_the_continental_scale	410658	CCME	aridisol	NA	NA	103	19502440	lauber_88_soils	NA	y	NA	SF1_V2	ENVO:soil	16S rRNA	NA	NA	V2	CCME	V2	NA	NA	Engencore	NA	35.38333333	NA	NA	soil metagenome
SF2.141677	AGCACACCTACA	CATGCTGCCTCCCGTAGGAGT	chris.lauber@gmail.com	NA	NA	ENVO:shrubland	Soil	source	NA	n	5/28/08	Pyrosequencing-Based Assessment of Soil pH as a Predictor of Soil Bacterial Community Structure at the Continental Scale	NA	FA6P1OK	NA	mimarks-survey	NA	NA	NA	4.2	0	NA	"Santa Fe, NM USA"	1500	2008	NA	0	NA	NA	NA	8.14	13	NA	NA	NA	SF2	1	"0.5, g"	NA	NA	targeted gene survey	4	NA	-105.9333333	y	7.4	CCME	NA	NA	NA	"Soils harbor enormously diverse bacterial populations, and soil bacterial communities can vary greatly in composition across space. However, our understanding of the specific changes in soil bacterial community structure that occur across larger spatial scales is limited because most previous work has focused on either surveying a relatively small number of soils in detail or analyzing a larger number of soils with techniques that provide little detail about the phylogenetic structure of the bacterial communities. Here we used a bar-coded pyrosequencing technique to characterize bacterial communities in 88 soils from across North and South America, obtaining an average of 1,501 sequences per soil. We found that overall bacterial community composition, as measured by pairwise UniFrac distances, was significantly correlated with differences in soil pH (r = 0.79), largely driven by changes in the relative abundances of Acidobacteria, Actinobacteria, and Bacteroidetes across the range of soil pHs. In addition, soil pH explains a significant portion of the variability associated with observed changes in the phylogenetic structure within each dominant lineage. The overall phylogenetic diversity of the bacterial communities was also correlated with soil pH (R2 = 0.50), with peak diversity in soils with near-neutral pHs. Together, these results suggest that the structure of soil bacterial communities is predictable, to some degree, across larger spatial scales, and the effect of soil pH on bacterial community composition is evident at even relatively coarse levels of taxonomic resolution."	ENVO:shrubland	TCAG	Pyrosequencing_based_assessment_of_soil_pH_as_a_predictor_of_soil_bacterial_community_structure_at_the_continental_scale	SF2_V2	0	bacterial biogeography	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19502440	NA	NA	NA	Chris Lauber	CCME	GAZ:United States of America	clay loam	lauber_88_soils	0-0.05	NA	NA	58	NA	n	NA	NA	Lauber_88_soils	pyrosequencing	NA	NA	494	soil metagenome	NA	ENVO:shrubland	FLX	SF2_V2	NA	250	8.38	NA	Lauber88Soils	Pyrosequencing_based_assessment_of_soil_pH_as_a_predictor_of_soil_bacterial_community_structure_at_the_continental_scale	410658	CCME	aridisol	NA	NA	103	19502440	lauber_88_soils	NA	y	NA	SF2_V2	ENVO:soil	16S rRNA	NA	NA	V2	CCME	V2	NA	NA	Engencore	NA	35.38333333	NA	NA	soil metagenome
SN3.141650	AGCGACTGTGCA	CATGCTGCCTCCCGTAGGAGT	chris.lauber@gmail.com	NA	NA	ENVO:shrubland	Soil	source	NA	n	5/28/08	Pyrosequencing-Based Assessment of Soil pH as a Predictor of Soil Bacterial Community Structure at the Continental Scale	NA	FA6P1OK	NA	mimarks-survey	NA	NA	NA	16.6	0	NA	"Sierra Nevada Mts., CA USA"	3000	2008	NA	0	NA	NA	NA	13.9	3.6	NA	NA	NA	SN3	1	"0.5, g"	NA	NA	targeted gene survey	6.289	NA	-118.1666667	y	0	CCME	NA	NA	NA	"Soils harbor enormously diverse bacterial populations, and soil bacterial communities can vary greatly in composition across space. However, our understanding of the specific changes in soil bacterial community structure that occur across larger spatial scales is limited because most previous work has focused on either surveying a relatively small number of soils in detail or analyzing a larger number of soils with techniques that provide little detail about the phylogenetic structure of the bacterial communities. Here we used a bar-coded pyrosequencing technique to characterize bacterial communities in 88 soils from across North and South America, obtaining an average of 1,501 sequences per soil. We found that overall bacterial community composition, as measured by pairwise UniFrac distances, was significantly correlated with differences in soil pH (r = 0.79), largely driven by changes in the relative abundances of Acidobacteria, Actinobacteria, and Bacteroidetes across the range of soil pHs. In addition, soil pH explains a significant portion of the variability associated with observed changes in the phylogenetic structure within each dominant lineage. The overall phylogenetic diversity of the bacterial communities was also correlated with soil pH (R2 = 0.50), with peak diversity in soils with near-neutral pHs. Together, these results suggest that the structure of soil bacterial communities is predictable, to some degree, across larger spatial scales, and the effect of soil pH on bacterial community composition is evident at even relatively coarse levels of taxonomic resolution."	ENVO:shrubland	TCAG	Pyrosequencing_based_assessment_of_soil_pH_as_a_predictor_of_soil_bacterial_community_structure_at_the_continental_scale	SN3_V2	0	bacterial biogeography	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19502440	NA	NA	NA	Chris Lauber	CCME	GAZ:United States of America	loamy sand	lauber_88_soils	0-0.05	NA	NA	20	NA	n	NA	NA	Lauber_88_soils	pyrosequencing	NA	NA	-252	soil metagenome	NA	ENVO:shrubland	FLX	SN3_V2	NA	600	5.74	NA	Lauber88Soils	Pyrosequencing_based_assessment_of_soil_pH_as_a_predictor_of_soil_bacterial_community_structure_at_the_continental_scale	410658	CCME	inceptisol	NA	NA	103	19502440	lauber_88_soils	NA	y	NA	SN3_V2	ENVO:soil	16S rRNA	NA	NA	V2	CCME	V2	NA	NA	Engencore	NA	36.45	NA	NA	soil metagenome
SP1.141656	AGCGAGCTATCT	CATGCTGCCTCCCGTAGGAGT	chris.lauber@gmail.com	NA	NA	ENVO:shrubland	Soil	source	NA	n	5/28/08	Pyrosequencing-Based Assessment of Soil pH as a Predictor of Soil Bacterial Community Structure at the Continental Scale	NA	FA6P1OK	NA	mimarks-survey	NA	NA	NA	16.8	0	NA	"Sequoia National Park, CA USA"	650	2008	NA	0	NA	NA	NA	18.5	12.7	NA	NA	NA	SP1	1	"0.5, g"	NA	NA	targeted gene survey	7.709	NA	-118.7	y	0	CCME	NA	NA	NA	"Soils harbor enormously diverse bacterial populations, and soil bacterial communities can vary greatly in composition across space. However, our understanding of the specific changes in soil bacterial community structure that occur across larger spatial scales is limited because most previous work has focused on either surveying a relatively small number of soils in detail or analyzing a larger number of soils with techniques that provide little detail about the phylogenetic structure of the bacterial communities. Here we used a bar-coded pyrosequencing technique to characterize bacterial communities in 88 soils from across North and South America, obtaining an average of 1,501 sequences per soil. We found that overall bacterial community composition, as measured by pairwise UniFrac distances, was significantly correlated with differences in soil pH (r = 0.79), largely driven by changes in the relative abundances of Acidobacteria, Actinobacteria, and Bacteroidetes across the range of soil pHs. In addition, soil pH explains a significant portion of the variability associated with observed changes in the phylogenetic structure within each dominant lineage. The overall phylogenetic diversity of the bacterial communities was also correlated with soil pH (R2 = 0.50), with peak diversity in soils with near-neutral pHs. Together, these results suggest that the structure of soil bacterial communities is predictable, to some degree, across larger spatial scales, and the effect of soil pH on bacterial community composition is evident at even relatively coarse levels of taxonomic resolution."	ENVO:shrubland	TCAG	Pyrosequencing_based_assessment_of_soil_pH_as_a_predictor_of_soil_bacterial_community_structure_at_the_continental_scale	SP1_V2	0	bacterial biogeography	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19502440	NA	NA	NA	Chris Lauber	CCME	GAZ:United States of America	sandy loam	lauber_88_soils	0-0.05	NA	NA	36	NA	n	NA	NA	Lauber_88_soils	pyrosequencing	NA	NA	17	soil metagenome	NA	ENVO:shrubland	FLX	SP1_V2	NA	650	6.25	NA	Lauber88Soils	Pyrosequencing_based_assessment_of_soil_pH_as_a_predictor_of_soil_bacterial_community_structure_at_the_continental_scale	410658	CCME	inceptisol	NA	NA	103	19502440	lauber_88_soils	NA	y	NA	SP1_V2	ENVO:soil	16S rRNA	NA	NA	V2	CCME	V2	NA	NA	Engencore	NA	36.5	NA	NA	soil metagenome
SR1.141680	AGCGTAGGTCGT	CATGCTGCCTCCCGTAGGAGT	chris.lauber@gmail.com	NA	NA	ENVO:shrubland	Soil	source	NA	n	5/28/08	Pyrosequencing-Based Assessment of Soil pH as a Predictor of Soil Bacterial Community Structure at the Continental Scale	NA	FA6P1OK	NA	mimarks-survey	NA	NA	NA	45.9	0	NA	"Sedgwick Reserve, CA USA"	300	2008	NA	0	NA	NA	NA	11.1	17.2	NA	NA	NA	SR1	1	"0.5, g"	NA	NA	targeted gene survey	6.119	NA	-120.05	y	0	CCME	NA	NA	NA	"Soils harbor enormously diverse bacterial populations, and soil bacterial communities can vary greatly in composition across space. However, our understanding of the specific changes in soil bacterial community structure that occur across larger spatial scales is limited because most previous work has focused on either surveying a relatively small number of soils in detail or analyzing a larger number of soils with techniques that provide little detail about the phylogenetic structure of the bacterial communities. Here we used a bar-coded pyrosequencing technique to characterize bacterial communities in 88 soils from across North and South America, obtaining an average of 1,501 sequences per soil. We found that overall bacterial community composition, as measured by pairwise UniFrac distances, was significantly correlated with differences in soil pH (r = 0.79), largely driven by changes in the relative abundances of Acidobacteria, Actinobacteria, and Bacteroidetes across the range of soil pHs. In addition, soil pH explains a significant portion of the variability associated with observed changes in the phylogenetic structure within each dominant lineage. The overall phylogenetic diversity of the bacterial communities was also correlated with soil pH (R2 = 0.50), with peak diversity in soils with near-neutral pHs. Together, these results suggest that the structure of soil bacterial communities is predictable, to some degree, across larger spatial scales, and the effect of soil pH on bacterial community composition is evident at even relatively coarse levels of taxonomic resolution."	ENVO:shrubland	TCAG	Pyrosequencing_based_assessment_of_soil_pH_as_a_predictor_of_soil_bacterial_community_structure_at_the_continental_scale	SR1_V2	0	bacterial biogeography	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19502440	NA	NA	NA	Chris Lauber	CCME	GAZ:United States of America	loam	lauber_88_soils	0-0.05	NA	NA	55	NA	n	NA	NA	Lauber_88_soils	pyrosequencing	NA	NA	324	soil metagenome	NA	ENVO:shrubland	FLX	SR1_V2	NA	500	6.84	NA	Lauber88Soils	Pyrosequencing_based_assessment_of_soil_pH_as_a_predictor_of_soil_bacterial_community_structure_at_the_continental_scale	410658	CCME	mollisol	NA	NA	103	19502440	lauber_88_soils	NA	y	NA	SR1_V2	ENVO:soil	16S rRNA	NA	NA	V2	CCME	V2	NA	NA	Engencore	NA	34.7	NA	NA	soil metagenome
SR3.141674	AGCTCCATACAG	CATGCTGCCTCCCGTAGGAGT	chris.lauber@gmail.com	NA	NA	ENVO:shrubland	Soil	source	NA	n	5/28/08	Pyrosequencing-Based Assessment of Soil pH as a Predictor of Soil Bacterial Community Structure at the Continental Scale	NA	FA6P1OK	NA	mimarks-survey	NA	NA	NA	33	0	NA	"Sedgwick Reserve, CA USA"	300	2008	NA	0	NA	NA	NA	11	17.2	NA	NA	NA	SR3	1	"0.5, g"	NA	NA	targeted gene survey	7.154	NA	-120.05	y	0	CCME	NA	NA	NA	"Soils harbor enormously diverse bacterial populations, and soil bacterial communities can vary greatly in composition across space. However, our understanding of the specific changes in soil bacterial community structure that occur across larger spatial scales is limited because most previous work has focused on either surveying a relatively small number of soils in detail or analyzing a larger number of soils with techniques that provide little detail about the phylogenetic structure of the bacterial communities. Here we used a bar-coded pyrosequencing technique to characterize bacterial communities in 88 soils from across North and South America, obtaining an average of 1,501 sequences per soil. We found that overall bacterial community composition, as measured by pairwise UniFrac distances, was significantly correlated with differences in soil pH (r = 0.79), largely driven by changes in the relative abundances of Acidobacteria, Actinobacteria, and Bacteroidetes across the range of soil pHs. In addition, soil pH explains a significant portion of the variability associated with observed changes in the phylogenetic structure within each dominant lineage. The overall phylogenetic diversity of the bacterial communities was also correlated with soil pH (R2 = 0.50), with peak diversity in soils with near-neutral pHs. Together, these results suggest that the structure of soil bacterial communities is predictable, to some degree, across larger spatial scales, and the effect of soil pH on bacterial community composition is evident at even relatively coarse levels of taxonomic resolution."	ENVO:shrubland	TCAG	Pyrosequencing_based_assessment_of_soil_pH_as_a_predictor_of_soil_bacterial_community_structure_at_the_continental_scale	SR3_V2	0	bacterial biogeography	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19502440	NA	NA	NA	Chris Lauber	CCME	GAZ:United States of America	loam	lauber_88_soils	0-0.05	NA	NA	61	NA	n	NA	NA	Lauber_88_soils	pyrosequencing	NA	NA	324	soil metagenome	NA	ENVO:shrubland	FLX	SR3_V2	NA	500	6.95	NA	Lauber88Soils	Pyrosequencing_based_assessment_of_soil_pH_as_a_predictor_of_soil_bacterial_community_structure_at_the_continental_scale	410658	CCME	mollisol	NA	NA	103	19502440	lauber_88_soils	NA	y	NA	SR3_V2	ENVO:soil	16S rRNA	NA	NA	V2	CCME	V2	NA	NA	Engencore	NA	34.68333333	NA	NA	soil metagenome
TL2.141706	AGGACGCACTGT	CATGCTGCCTCCCGTAGGAGT	chris.lauber@gmail.com	NA	NA	ENVO:shrubland	Soil	source	NA	n	5/28/08	Pyrosequencing-Based Assessment of Soil pH as a Predictor of Soil Bacterial Community Structure at the Continental Scale	NA	FA6P1OK	NA	mimarks-survey	NA	NA	NA	158.3	0	NA	"Toolik Lake LTER, AK USA"	894	2008	NA	0	NA	NA	NA	16.6	-9.3	NA	NA	NA	TL2	1	"0.5, g"	NA	NA	targeted gene survey	12.092	NA	-149.5833333	y	0	CCME	NA	NA	NA	"Soils harbor enormously diverse bacterial populations, and soil bacterial communities can vary greatly in composition across space. However, our understanding of the specific changes in soil bacterial community structure that occur across larger spatial scales is limited because most previous work has focused on either surveying a relatively small number of soils in detail or analyzing a larger number of soils with techniques that provide little detail about the phylogenetic structure of the bacterial communities. Here we used a bar-coded pyrosequencing technique to characterize bacterial communities in 88 soils from across North and South America, obtaining an average of 1,501 sequences per soil. We found that overall bacterial community composition, as measured by pairwise UniFrac distances, was significantly correlated with differences in soil pH (r = 0.79), largely driven by changes in the relative abundances of Acidobacteria, Actinobacteria, and Bacteroidetes across the range of soil pHs. In addition, soil pH explains a significant portion of the variability associated with observed changes in the phylogenetic structure within each dominant lineage. The overall phylogenetic diversity of the bacterial communities was also correlated with soil pH (R2 = 0.50), with peak diversity in soils with near-neutral pHs. Together, these results suggest that the structure of soil bacterial communities is predictable, to some degree, across larger spatial scales, and the effect of soil pH on bacterial community composition is evident at even relatively coarse levels of taxonomic resolution."	ENVO:shrubland	TCAG	Pyrosequencing_based_assessment_of_soil_pH_as_a_predictor_of_soil_bacterial_community_structure_at_the_continental_scale	TL2_V2	0	bacterial biogeography	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19502440	NA	NA	NA	Chris Lauber	CCME	GAZ:United States of America	silt loam	lauber_88_soils	0-0.05	NA	NA	61	NA	n	NA	NA	Lauber_88_soils	pyrosequencing	NA	NA	-212	soil metagenome	NA	ENVO:shrubland	FLX	TL2_V2	NA	400	6.47	NA	Lauber88Soils	Pyrosequencing_based_assessment_of_soil_pH_as_a_predictor_of_soil_bacterial_community_structure_at_the_continental_scale	410658	CCME	gelisol	NA	NA	103	19502440	lauber_88_soils	NA	y	NA	TL2_V2	ENVO:soil	16S rRNA	NA	NA	V2	CCME	V2	NA	NA	Engencore	NA	68.63333333	NA	NA	soil metagenome
TL3.141709	AGGCTACACGAC	CATGCTGCCTCCCGTAGGAGT	chris.lauber@gmail.com	NA	NA	ENVO:shrubland	Soil	source	NA	n	5/28/08	Pyrosequencing-Based Assessment of Soil pH as a Predictor of Soil Bacterial Community Structure at the Continental Scale	NA	FA6P1OK	NA	mimarks-survey	NA	NA	NA	53.9	0	NA	"Toolik Lake LTER, AK USA"	894	2008	NA	0	NA	NA	NA	24.6	-9.3	NA	NA	NA	TL3	1	"0.5, g"	NA	NA	targeted gene survey	16.456	NA	-149.5833333	y	0	CCME	NA	NA	NA	"Soils harbor enormously diverse bacterial populations, and soil bacterial communities can vary greatly in composition across space. However, our understanding of the specific changes in soil bacterial community structure that occur across larger spatial scales is limited because most previous work has focused on either surveying a relatively small number of soils in detail or analyzing a larger number of soils with techniques that provide little detail about the phylogenetic structure of the bacterial communities. Here we used a bar-coded pyrosequencing technique to characterize bacterial communities in 88 soils from across North and South America, obtaining an average of 1,501 sequences per soil. We found that overall bacterial community composition, as measured by pairwise UniFrac distances, was significantly correlated with differences in soil pH (r = 0.79), largely driven by changes in the relative abundances of Acidobacteria, Actinobacteria, and Bacteroidetes across the range of soil pHs. In addition, soil pH explains a significant portion of the variability associated with observed changes in the phylogenetic structure within each dominant lineage. The overall phylogenetic diversity of the bacterial communities was also correlated with soil pH (R2 = 0.50), with peak diversity in soils with near-neutral pHs. Together, these results suggest that the structure of soil bacterial communities is predictable, to some degree, across larger spatial scales, and the effect of soil pH on bacterial community composition is evident at even relatively coarse levels of taxonomic resolution."	ENVO:shrubland	TCAG	Pyrosequencing_based_assessment_of_soil_pH_as_a_predictor_of_soil_bacterial_community_structure_at_the_continental_scale	TL3_V2	0	bacterial biogeography	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19502440	NA	NA	NA	Chris Lauber	CCME	GAZ:United States of America	loam	lauber_88_soils	0-0.05	NA	NA	52	NA	n	NA	NA	Lauber_88_soils	pyrosequencing	NA	NA	-212	soil metagenome	NA	ENVO:shrubland	FLX	TL3_V2	NA	400	4.23	NA	Lauber88Soils	Pyrosequencing_based_assessment_of_soil_pH_as_a_predictor_of_soil_bacterial_community_structure_at_the_continental_scale	410658	CCME	gelisol	NA	NA	103	19502440	lauber_88_soils	NA	y	NA	TL3_V2	ENVO:soil	16S rRNA	NA	NA	V2	CCME	V2	NA	NA	Engencore	NA	68.63333333	NA	NA	soil metagenome
AR1.141727	AACGCACGCTAG	CATGCTGCCTCCCGTAGGAGT	chris.lauber@gmail.com	NA	NA	ENVO:tropical_soil	Soil	source	NA	n	5/28/08	Pyrosequencing-Based Assessment of Soil pH as a Predictor of Soil Bacterial Community Structure at the Continental Scale	NA	FA6P1OK	NA	mimarks-survey	NA	NA	NA	22.1	0	NA	"Misiones, Argentina"	150	2008	NA	0	NA	NA	NA	8.98	23	NA	NA	NA	AR1	1	"0.5, g"	NA	NA	targeted gene survey	4.3	NA	-55.68333333	y	0	CCME	NA	NA	NA	"Soils harbor enormously diverse bacterial populations, and soil bacterial communities can vary greatly in composition across space. However, our understanding of the specific changes in soil bacterial community structure that occur across larger spatial scales is limited because most previous work has focused on either surveying a relatively small number of soils in detail or analyzing a larger number of soils with techniques that provide little detail about the phylogenetic structure of the bacterial communities. Here we used a bar-coded pyrosequencing technique to characterize bacterial communities in 88 soils from across North and South America, obtaining an average of 1,501 sequences per soil. We found that overall bacterial community composition, as measured by pairwise UniFrac distances, was significantly correlated with differences in soil pH (r = 0.79), largely driven by changes in the relative abundances of Acidobacteria, Actinobacteria, and Bacteroidetes across the range of soil pHs. In addition, soil pH explains a significant portion of the variability associated with observed changes in the phylogenetic structure within each dominant lineage. The overall phylogenetic diversity of the bacterial communities was also correlated with soil pH (R2 = 0.50), with peak diversity in soils with near-neutral pHs. Together, these results suggest that the structure of soil bacterial communities is predictable, to some degree, across larger spatial scales, and the effect of soil pH on bacterial community composition is evident at even relatively coarse levels of taxonomic resolution."	ENVO:tropical soil	TCAG	Pyrosequencing_based_assessment_of_soil_pH_as_a_predictor_of_soil_bacterial_community_structure_at_the_continental_scale	AR1_V2	0	bacterial biogeography	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19502440	NA	NA	NA	Chris Lauber	CCME	GAZ:Argentina	clay	lauber_88_soils	0-0.05	NA	NA	80	NA	n	NA	NA	Lauber_88_soils	pyrosequencing	NA	NA	-206	soil metagenome	NA	ENVO:Tropical humid forests	FLX	AR1_V2	NA	1400	5.8	NA	Lauber88Soils	Pyrosequencing_based_assessment_of_soil_pH_as_a_predictor_of_soil_bacterial_community_structure_at_the_continental_scale	410658	CCME	oxisol	NA	NA	103	19502440	lauber_88_soils	NA	y	NA	AR1_V2	ENVO:soil	16S rRNA	NA	NA	V2	CCME	V2	NA	NA	Engencore	NA	-27.73333333	NA	NA	soil metagenome
AR2.141703	AACTCGTCGATG	CATGCTGCCTCCCGTAGGAGT	chris.lauber@gmail.com	NA	NA	ENVO:tropical_soil	Soil	source	NA	n	5/28/08	Pyrosequencing-Based Assessment of Soil pH as a Predictor of Soil Bacterial Community Structure at the Continental Scale	NA	FA6P1OK	NA	mimarks-survey	NA	NA	NA	21.7	0	NA	"Misiones, Argentina"	150	2008	NA	0	NA	NA	NA	9.36	23	NA	NA	NA	AR2	1	"0.5, g"	NA	NA	targeted gene survey	3.9	NA	-55.68333333	y	0	CCME	NA	NA	NA	"Soils harbor enormously diverse bacterial populations, and soil bacterial communities can vary greatly in composition across space. However, our understanding of the specific changes in soil bacterial community structure that occur across larger spatial scales is limited because most previous work has focused on either surveying a relatively small number of soils in detail or analyzing a larger number of soils with techniques that provide little detail about the phylogenetic structure of the bacterial communities. Here we used a bar-coded pyrosequencing technique to characterize bacterial communities in 88 soils from across North and South America, obtaining an average of 1,501 sequences per soil. We found that overall bacterial community composition, as measured by pairwise UniFrac distances, was significantly correlated with differences in soil pH (r = 0.79), largely driven by changes in the relative abundances of Acidobacteria, Actinobacteria, and Bacteroidetes across the range of soil pHs. In addition, soil pH explains a significant portion of the variability associated with observed changes in the phylogenetic structure within each dominant lineage. The overall phylogenetic diversity of the bacterial communities was also correlated with soil pH (R2 = 0.50), with peak diversity in soils with near-neutral pHs. Together, these results suggest that the structure of soil bacterial communities is predictable, to some degree, across larger spatial scales, and the effect of soil pH on bacterial community composition is evident at even relatively coarse levels of taxonomic resolution."	ENVO:tropical soil	TCAG	Pyrosequencing_based_assessment_of_soil_pH_as_a_predictor_of_soil_bacterial_community_structure_at_the_continental_scale	AR2_V2	0	bacterial biogeography	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19502440	NA	NA	NA	Chris Lauber	CCME	GAZ:Argentina	clay	lauber_88_soils	0-0.05	NA	NA	78	NA	n	NA	NA	Lauber_88_soils	pyrosequencing	NA	NA	-206	soil metagenome	NA	ENVO:Tropical humid forests	FLX	AR2_V2	NA	1400	6	NA	Lauber88Soils	Pyrosequencing_based_assessment_of_soil_pH_as_a_predictor_of_soil_bacterial_community_structure_at_the_continental_scale	410658	CCME	oxisol	NA	NA	103	19502440	lauber_88_soils	NA	y	NA	AR2_V2	ENVO:soil	16S rRNA	NA	NA	V2	CCME	V2	NA	NA	Engencore	NA	-27.73333333	NA	NA	soil metagenome
AR3.141705	AACTGTGCGTAC	CATGCTGCCTCCCGTAGGAGT	chris.lauber@gmail.com	NA	NA	ENVO:tropical_soil	Soil	source	NA	n	5/28/08	Pyrosequencing-Based Assessment of Soil pH as a Predictor of Soil Bacterial Community Structure at the Continental Scale	NA	FA6P1OK	NA	mimarks-survey	NA	NA	NA	24.4	0	NA	"Misiones, Argentina"	150	2008	NA	0	NA	NA	NA	9.844	23	NA	NA	NA	AR3	1	"0.5, g"	NA	NA	targeted gene survey	4.5	NA	-55.68333333	y	0	CCME	NA	NA	NA	"Soils harbor enormously diverse bacterial populations, and soil bacterial communities can vary greatly in composition across space. However, our understanding of the specific changes in soil bacterial community structure that occur across larger spatial scales is limited because most previous work has focused on either surveying a relatively small number of soils in detail or analyzing a larger number of soils with techniques that provide little detail about the phylogenetic structure of the bacterial communities. Here we used a bar-coded pyrosequencing technique to characterize bacterial communities in 88 soils from across North and South America, obtaining an average of 1,501 sequences per soil. We found that overall bacterial community composition, as measured by pairwise UniFrac distances, was significantly correlated with differences in soil pH (r = 0.79), largely driven by changes in the relative abundances of Acidobacteria, Actinobacteria, and Bacteroidetes across the range of soil pHs. In addition, soil pH explains a significant portion of the variability associated with observed changes in the phylogenetic structure within each dominant lineage. The overall phylogenetic diversity of the bacterial communities was also correlated with soil pH (R2 = 0.50), with peak diversity in soils with near-neutral pHs. Together, these results suggest that the structure of soil bacterial communities is predictable, to some degree, across larger spatial scales, and the effect of soil pH on bacterial community composition is evident at even relatively coarse levels of taxonomic resolution."	ENVO:tropical soil	TCAG	Pyrosequencing_based_assessment_of_soil_pH_as_a_predictor_of_soil_bacterial_community_structure_at_the_continental_scale	AR3_V2	0	bacterial biogeography	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19502440	NA	NA	NA	Chris Lauber	CCME	GAZ:Argentina	clay	lauber_88_soils	0-0.05	NA	NA	82	NA	n	NA	NA	Lauber_88_soils	pyrosequencing	NA	NA	-206	soil metagenome	NA	ENVO:Tropical humid forests	FLX	AR3_V2	NA	1400	5.9	NA	Lauber88Soils	Pyrosequencing_based_assessment_of_soil_pH_as_a_predictor_of_soil_bacterial_community_structure_at_the_continental_scale	410658	CCME	oxisol	NA	NA	103	19502440	lauber_88_soils	NA	y	NA	AR3_V2	ENVO:soil	16S rRNA	NA	NA	V2	CCME	V2	NA	NA	Engencore	NA	-27.73333333	NA	NA	soil metagenome
LQ2.141729	ACTCACGGTATG	CATGCTGCCTCCCGTAGGAGT	chris.lauber@gmail.com	NA	NA	ENVO:tropical_soil	Soil	source	NA	n	5/28/08	Pyrosequencing-Based Assessment of Soil pH as a Predictor of Soil Bacterial Community Structure at the Continental Scale	NA	FA6P1OK	NA	mimarks-survey	NA	NA	NA	41.1	0	NA	"Luquillo LTER, Puerto Rico"	400	2008	NA	0	NA	NA	NA	13	21.5	NA	NA	NA	LQ2	1	"0.5, g"	NA	NA	targeted gene survey	7.816	NA	-65.83333333	y	0	CCME	NA	NA	NA	"Soils harbor enormously diverse bacterial populations, and soil bacterial communities can vary greatly in composition across space. However, our understanding of the specific changes in soil bacterial community structure that occur across larger spatial scales is limited because most previous work has focused on either surveying a relatively small number of soils in detail or analyzing a larger number of soils with techniques that provide little detail about the phylogenetic structure of the bacterial communities. Here we used a bar-coded pyrosequencing technique to characterize bacterial communities in 88 soils from across North and South America, obtaining an average of 1,501 sequences per soil. We found that overall bacterial community composition, as measured by pairwise UniFrac distances, was significantly correlated with differences in soil pH (r = 0.79), largely driven by changes in the relative abundances of Acidobacteria, Actinobacteria, and Bacteroidetes across the range of soil pHs. In addition, soil pH explains a significant portion of the variability associated with observed changes in the phylogenetic structure within each dominant lineage. The overall phylogenetic diversity of the bacterial communities was also correlated with soil pH (R2 = 0.50), with peak diversity in soils with near-neutral pHs. Together, these results suggest that the structure of soil bacterial communities is predictable, to some degree, across larger spatial scales, and the effect of soil pH on bacterial community composition is evident at even relatively coarse levels of taxonomic resolution."	ENVO:tropical soil	TCAG	Pyrosequencing_based_assessment_of_soil_pH_as_a_predictor_of_soil_bacterial_community_structure_at_the_continental_scale	LQ2_V2	0	bacterial biogeography	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19502440	NA	NA	NA	Chris Lauber	CCME	GAZ:Puerto Rico	silty clay loam	lauber_88_soils	0-0.05	NA	NA	84	NA	n	NA	NA	Lauber_88_soils	pyrosequencing	NA	NA	-2454	soil metagenome	NA	ENVO:Tropical humid forests	FLX	LQ2_V2	NA	3500	5.03	NA	Lauber88Soils	Pyrosequencing_based_assessment_of_soil_pH_as_a_predictor_of_soil_bacterial_community_structure_at_the_continental_scale	410658	CCME	inceptisol	NA	NA	103	19502440	lauber_88_soils	NA	y	NA	LQ2_V2	ENVO:soil	16S rRNA	NA	NA	V2	CCME	V2	NA	NA	Engencore	NA	18.3	NA	NA	soil metagenome
LQ3.141712	ACTCAGATACTC	CATGCTGCCTCCCGTAGGAGT	chris.lauber@gmail.com	NA	NA	ENVO:tropical_soil	Soil	source	NA	n	5/28/08	Pyrosequencing-Based Assessment of Soil pH as a Predictor of Soil Bacterial Community Structure at the Continental Scale	NA	FA6P1OK	NA	mimarks-survey	NA	NA	NA	64.1	0	NA	"Luquillo LTER, Puerto Rico"	700	2008	NA	0	NA	NA	NA	22.6	20.5	NA	NA	NA	LQ3	1	"0.5, g"	NA	NA	targeted gene survey	3.686	NA	-65.83333333	y	0	CCME	NA	NA	NA	"Soils harbor enormously diverse bacterial populations, and soil bacterial communities can vary greatly in composition across space. However, our understanding of the specific changes in soil bacterial community structure that occur across larger spatial scales is limited because most previous work has focused on either surveying a relatively small number of soils in detail or analyzing a larger number of soils with techniques that provide little detail about the phylogenetic structure of the bacterial communities. Here we used a bar-coded pyrosequencing technique to characterize bacterial communities in 88 soils from across North and South America, obtaining an average of 1,501 sequences per soil. We found that overall bacterial community composition, as measured by pairwise UniFrac distances, was significantly correlated with differences in soil pH (r = 0.79), largely driven by changes in the relative abundances of Acidobacteria, Actinobacteria, and Bacteroidetes across the range of soil pHs. In addition, soil pH explains a significant portion of the variability associated with observed changes in the phylogenetic structure within each dominant lineage. The overall phylogenetic diversity of the bacterial communities was also correlated with soil pH (R2 = 0.50), with peak diversity in soils with near-neutral pHs. Together, these results suggest that the structure of soil bacterial communities is predictable, to some degree, across larger spatial scales, and the effect of soil pH on bacterial community composition is evident at even relatively coarse levels of taxonomic resolution."	ENVO:tropical soil	TCAG	Pyrosequencing_based_assessment_of_soil_pH_as_a_predictor_of_soil_bacterial_community_structure_at_the_continental_scale	LQ3_V2	0	bacterial biogeography	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19502440	NA	NA	NA	Chris Lauber	CCME	GAZ:Puerto Rico	sandy loam	lauber_88_soils	0-0.05	NA	NA	35	NA	n	NA	NA	Lauber_88_soils	pyrosequencing	NA	NA	-3528	soil metagenome	NA	ENVO:Tropical humid forests	FLX	LQ3_V2	NA	4500	4.67	NA	Lauber88Soils	Pyrosequencing_based_assessment_of_soil_pH_as_a_predictor_of_soil_bacterial_community_structure_at_the_continental_scale	410658	CCME	inceptisol	NA	NA	103	19502440	lauber_88_soils	NA	y	NA	LQ3_V2	ENVO:soil	16S rRNA	NA	NA	V2	CCME	V2	NA	NA	Engencore	NA	18.3	NA	NA	soil metagenome
PE1.141715	ACTTGTAGCAGC	CATGCTGCCTCCCGTAGGAGT	chris.lauber@gmail.com	NA	NA	ENVO:tropical_soil	Soil	source	NA	n	5/28/08	Pyrosequencing-Based Assessment of Soil pH as a Predictor of Soil Bacterial Community Structure at the Continental Scale	NA	FA6P1OK	NA	mimarks-survey	NA	NA	NA	134.2	0	NA	"Manu National Park, Peru"	3250	2008	NA	0	NA	NA	NA	13.887	10	NA	NA	NA	PE1	1	"0.5, g"	NA	NA	targeted gene survey	32.2	NA	-71.58333333	y	0	CCME	NA	NA	NA	"Soils harbor enormously diverse bacterial populations, and soil bacterial communities can vary greatly in composition across space. However, our understanding of the specific changes in soil bacterial community structure that occur across larger spatial scales is limited because most previous work has focused on either surveying a relatively small number of soils in detail or analyzing a larger number of soils with techniques that provide little detail about the phylogenetic structure of the bacterial communities. Here we used a bar-coded pyrosequencing technique to characterize bacterial communities in 88 soils from across North and South America, obtaining an average of 1,501 sequences per soil. We found that overall bacterial community composition, as measured by pairwise UniFrac distances, was significantly correlated with differences in soil pH (r = 0.79), largely driven by changes in the relative abundances of Acidobacteria, Actinobacteria, and Bacteroidetes across the range of soil pHs. In addition, soil pH explains a significant portion of the variability associated with observed changes in the phylogenetic structure within each dominant lineage. The overall phylogenetic diversity of the bacterial communities was also correlated with soil pH (R2 = 0.50), with peak diversity in soils with near-neutral pHs. Together, these results suggest that the structure of soil bacterial communities is predictable, to some degree, across larger spatial scales, and the effect of soil pH on bacterial community composition is evident at even relatively coarse levels of taxonomic resolution."	ENVO:tropical soil	TCAG	Pyrosequencing_based_assessment_of_soil_pH_as_a_predictor_of_soil_bacterial_community_structure_at_the_continental_scale	PE1_V2	0	bacterial biogeography	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19502440	NA	NA	NA	Chris Lauber	CCME	GAZ:Peru	sandy loam	lauber_88_soils	0-0.05	NA	NA	35	NA	n	NA	NA	Lauber_88_soils	pyrosequencing	NA	NA	-2500	soil metagenome	NA	ENVO:Tropical humid forests	FLX	PE1_V2	NA	2100	4.12	NA	Lauber88Soils	Pyrosequencing_based_assessment_of_soil_pH_as_a_predictor_of_soil_bacterial_community_structure_at_the_continental_scale	410658	CCME	inceptisol	NA	NA	103	19502440	lauber_88_soils	NA	y	NA	PE1_V2	ENVO:soil	16S rRNA	NA	NA	V2	CCME	V2	NA	NA	Engencore	NA	-13.08333333	NA	NA	soil metagenome
PE3.141731	AGACCGTCAGAC	CATGCTGCCTCCCGTAGGAGT	chris.lauber@gmail.com	NA	NA	ENVO:tropical_soil	Soil	source	NA	n	5/28/08	Pyrosequencing-Based Assessment of Soil pH as a Predictor of Soil Bacterial Community Structure at the Continental Scale	NA	FA6P1OK	NA	mimarks-survey	NA	NA	NA	130.9	0	NA	"Manu National Park, Peru"	2750	2008	NA	0	NA	NA	NA	16.84	16	NA	NA	NA	PE3	1	"0.5, g"	NA	NA	targeted gene survey	31.8	NA	-71.58333333	y	0	CCME	NA	NA	NA	"Soils harbor enormously diverse bacterial populations, and soil bacterial communities can vary greatly in composition across space. However, our understanding of the specific changes in soil bacterial community structure that occur across larger spatial scales is limited because most previous work has focused on either surveying a relatively small number of soils in detail or analyzing a larger number of soils with techniques that provide little detail about the phylogenetic structure of the bacterial communities. Here we used a bar-coded pyrosequencing technique to characterize bacterial communities in 88 soils from across North and South America, obtaining an average of 1,501 sequences per soil. We found that overall bacterial community composition, as measured by pairwise UniFrac distances, was significantly correlated with differences in soil pH (r = 0.79), largely driven by changes in the relative abundances of Acidobacteria, Actinobacteria, and Bacteroidetes across the range of soil pHs. In addition, soil pH explains a significant portion of the variability associated with observed changes in the phylogenetic structure within each dominant lineage. The overall phylogenetic diversity of the bacterial communities was also correlated with soil pH (R2 = 0.50), with peak diversity in soils with near-neutral pHs. Together, these results suggest that the structure of soil bacterial communities is predictable, to some degree, across larger spatial scales, and the effect of soil pH on bacterial community composition is evident at even relatively coarse levels of taxonomic resolution."	ENVO:tropical soil	TCAG	Pyrosequencing_based_assessment_of_soil_pH_as_a_predictor_of_soil_bacterial_community_structure_at_the_continental_scale	PE3_V2	0	bacterial biogeography	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19502440	NA	NA	NA	Chris Lauber	CCME	GAZ:Peru	sandy loam	lauber_88_soils	0-0.05	NA	NA	40	NA	n	NA	NA	Lauber_88_soils	pyrosequencing	NA	NA	-3270	soil metagenome	NA	ENVO:Tropical humid forests	FLX	PE3_V2	NA	5500	4.25	NA	Lauber88Soils	Pyrosequencing_based_assessment_of_soil_pH_as_a_predictor_of_soil_bacterial_community_structure_at_the_continental_scale	410658	CCME	inceptisol	NA	NA	103	19502440	lauber_88_soils	NA	y	NA	PE3_V2	ENVO:soil	16S rRNA	NA	NA	V2	CCME	V2	NA	NA	Engencore	NA	-13.08333333	NA	NA	soil metagenome
PE4.141683	AGACGTGCACTG	CATGCTGCCTCCCGTAGGAGT	chris.lauber@gmail.com	NA	NA	ENVO:tropical_soil	Soil	source	NA	n	5/28/08	Pyrosequencing-Based Assessment of Soil pH as a Predictor of Soil Bacterial Community Structure at the Continental Scale	NA	FA6P1OK	NA	mimarks-survey	NA	NA	NA	59.5	0	NA	"Manu National Park, Peru"	2000	2008	NA	0	NA	NA	NA	18.45	17	NA	NA	NA	PE4	1	"0.5, g"	NA	NA	targeted gene survey	15.5	NA	-71.58333333	y	0	CCME	NA	NA	NA	"Soils harbor enormously diverse bacterial populations, and soil bacterial communities can vary greatly in composition across space. However, our understanding of the specific changes in soil bacterial community structure that occur across larger spatial scales is limited because most previous work has focused on either surveying a relatively small number of soils in detail or analyzing a larger number of soils with techniques that provide little detail about the phylogenetic structure of the bacterial communities. Here we used a bar-coded pyrosequencing technique to characterize bacterial communities in 88 soils from across North and South America, obtaining an average of 1,501 sequences per soil. We found that overall bacterial community composition, as measured by pairwise UniFrac distances, was significantly correlated with differences in soil pH (r = 0.79), largely driven by changes in the relative abundances of Acidobacteria, Actinobacteria, and Bacteroidetes across the range of soil pHs. In addition, soil pH explains a significant portion of the variability associated with observed changes in the phylogenetic structure within each dominant lineage. The overall phylogenetic diversity of the bacterial communities was also correlated with soil pH (R2 = 0.50), with peak diversity in soils with near-neutral pHs. Together, these results suggest that the structure of soil bacterial communities is predictable, to some degree, across larger spatial scales, and the effect of soil pH on bacterial community composition is evident at even relatively coarse levels of taxonomic resolution."	ENVO:tropical soil	TCAG	Pyrosequencing_based_assessment_of_soil_pH_as_a_predictor_of_soil_bacterial_community_structure_at_the_continental_scale	PE4_V2	0	bacterial biogeography	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19502440	NA	NA	NA	Chris Lauber	CCME	GAZ:Peru	sandy loam	lauber_88_soils	0-0.05	NA	NA	32	NA	n	NA	NA	Lauber_88_soils	pyrosequencing	NA	NA	-3700	soil metagenome	NA	ENVO:Tropical humid forests	FLX	PE4_V2	NA	6200	4.1	NA	Lauber88Soils	Pyrosequencing_based_assessment_of_soil_pH_as_a_predictor_of_soil_bacterial_community_structure_at_the_continental_scale	410658	CCME	inceptisol	NA	NA	103	19502440	lauber_88_soils	NA	y	NA	PE4_V2	ENVO:soil	16S rRNA	NA	NA	V2	CCME	V2	NA	NA	Engencore	NA	-13.08333333	NA	NA	soil metagenome
PE5.141692	AGACTGCGTACT	CATGCTGCCTCCCGTAGGAGT	chris.lauber@gmail.com	NA	NA	ENVO:tropical_soil	Soil	source	NA	n	5/28/08	Pyrosequencing-Based Assessment of Soil pH as a Predictor of Soil Bacterial Community Structure at the Continental Scale	NA	FA6P1OK	NA	mimarks-survey	NA	NA	NA	93.6	0	NA	"Manu National Park, Peru"	1750	2008	NA	0	NA	NA	NA	14.55	23	NA	NA	NA	PE5	1	"0.5, g"	NA	NA	targeted gene survey	25	NA	-71.26666667	y	0	CCME	NA	NA	NA	"Soils harbor enormously diverse bacterial populations, and soil bacterial communities can vary greatly in composition across space. However, our understanding of the specific changes in soil bacterial community structure that occur across larger spatial scales is limited because most previous work has focused on either surveying a relatively small number of soils in detail or analyzing a larger number of soils with techniques that provide little detail about the phylogenetic structure of the bacterial communities. Here we used a bar-coded pyrosequencing technique to characterize bacterial communities in 88 soils from across North and South America, obtaining an average of 1,501 sequences per soil. We found that overall bacterial community composition, as measured by pairwise UniFrac distances, was significantly correlated with differences in soil pH (r = 0.79), largely driven by changes in the relative abundances of Acidobacteria, Actinobacteria, and Bacteroidetes across the range of soil pHs. In addition, soil pH explains a significant portion of the variability associated with observed changes in the phylogenetic structure within each dominant lineage. The overall phylogenetic diversity of the bacterial communities was also correlated with soil pH (R2 = 0.50), with peak diversity in soils with near-neutral pHs. Together, these results suggest that the structure of soil bacterial communities is predictable, to some degree, across larger spatial scales, and the effect of soil pH on bacterial community composition is evident at even relatively coarse levels of taxonomic resolution."	ENVO:tropical soil	TCAG	Pyrosequencing_based_assessment_of_soil_pH_as_a_predictor_of_soil_bacterial_community_structure_at_the_continental_scale	PE5_V2	0	bacterial biogeography	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19502440	NA	NA	NA	Chris Lauber	CCME	GAZ:Peru	clay loam	lauber_88_soils	0-0.05	NA	NA	70	NA	n	NA	NA	Lauber_88_soils	pyrosequencing	NA	NA	-1900	soil metagenome	NA	ENVO:Tropical humid forests	FLX	PE5_V2	NA	5000	3.57	NA	Lauber88Soils	Pyrosequencing_based_assessment_of_soil_pH_as_a_predictor_of_soil_bacterial_community_structure_at_the_continental_scale	410658	CCME	oxisol	NA	NA	103	19502440	lauber_88_soils	NA	y	NA	PE5_V2	ENVO:soil	16S rRNA	NA	NA	V2	CCME	V2	NA	NA	Engencore	NA	-12.63333333	NA	NA	soil metagenome
PE6.141700	AGAGAGCAAGTG	CATGCTGCCTCCCGTAGGAGT	chris.lauber@gmail.com	NA	NA	ENVO:tropical_soil	Soil	source	NA	n	5/28/08	Pyrosequencing-Based Assessment of Soil pH as a Predictor of Soil Bacterial Community Structure at the Continental Scale	NA	FA6P1OK	NA	mimarks-survey	NA	NA	NA	33.4	0	NA	"Manu National Park, Peru"	860	2008	NA	0	NA	NA	NA	10.12	25	NA	NA	NA	PE6	1	"0.5, g"	NA	NA	targeted gene survey	13.4	NA	-71.23333333	y	0	CCME	NA	NA	NA	"Soils harbor enormously diverse bacterial populations, and soil bacterial communities can vary greatly in composition across space. However, our understanding of the specific changes in soil bacterial community structure that occur across larger spatial scales is limited because most previous work has focused on either surveying a relatively small number of soils in detail or analyzing a larger number of soils with techniques that provide little detail about the phylogenetic structure of the bacterial communities. Here we used a bar-coded pyrosequencing technique to characterize bacterial communities in 88 soils from across North and South America, obtaining an average of 1,501 sequences per soil. We found that overall bacterial community composition, as measured by pairwise UniFrac distances, was significantly correlated with differences in soil pH (r = 0.79), largely driven by changes in the relative abundances of Acidobacteria, Actinobacteria, and Bacteroidetes across the range of soil pHs. In addition, soil pH explains a significant portion of the variability associated with observed changes in the phylogenetic structure within each dominant lineage. The overall phylogenetic diversity of the bacterial communities was also correlated with soil pH (R2 = 0.50), with peak diversity in soils with near-neutral pHs. Together, these results suggest that the structure of soil bacterial communities is predictable, to some degree, across larger spatial scales, and the effect of soil pH on bacterial community composition is evident at even relatively coarse levels of taxonomic resolution."	ENVO:tropical soil	TCAG	Pyrosequencing_based_assessment_of_soil_pH_as_a_predictor_of_soil_bacterial_community_structure_at_the_continental_scale	PE6_V2	0	bacterial biogeography	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19502440	NA	NA	NA	Chris Lauber	CCME	GAZ:Peru	clay	lauber_88_soils	0-0.05	NA	NA	65	NA	n	NA	NA	Lauber_88_soils	pyrosequencing	NA	NA	-1600	soil metagenome	NA	ENVO:Tropical humid forests	FLX	PE6_V2	NA	4000	4.12	NA	Lauber88Soils	Pyrosequencing_based_assessment_of_soil_pH_as_a_predictor_of_soil_bacterial_community_structure_at_the_continental_scale	410658	CCME	oxisol	NA	NA	103	19502440	lauber_88_soils	NA	y	NA	PE6_V2	ENVO:soil	16S rRNA	NA	NA	V2	CCME	V2	NA	NA	Engencore	NA	-12.65	NA	NA	soil metagenome
PE7.141734	AGAGCAAGAGCA	CATGCTGCCTCCCGTAGGAGT	chris.lauber@gmail.com	NA	NA	ENVO:tropical_soil	Soil	source	NA	n	5/28/08	Pyrosequencing-Based Assessment of Soil pH as a Predictor of Soil Bacterial Community Structure at the Continental Scale	NA	FA6P1OK	NA	mimarks-survey	NA	NA	NA	63.8	0	NA	"Manu National Park, Peru"	440	2008	NA	0	NA	NA	NA	11.21	25	NA	NA	NA	PE7	1	"0.5, g"	NA	NA	targeted gene survey	16.7	NA	-71.23333333	y	0	CCME	NA	NA	NA	"Soils harbor enormously diverse bacterial populations, and soil bacterial communities can vary greatly in composition across space. However, our understanding of the specific changes in soil bacterial community structure that occur across larger spatial scales is limited because most previous work has focused on either surveying a relatively small number of soils in detail or analyzing a larger number of soils with techniques that provide little detail about the phylogenetic structure of the bacterial communities. Here we used a bar-coded pyrosequencing technique to characterize bacterial communities in 88 soils from across North and South America, obtaining an average of 1,501 sequences per soil. We found that overall bacterial community composition, as measured by pairwise UniFrac distances, was significantly correlated with differences in soil pH (r = 0.79), largely driven by changes in the relative abundances of Acidobacteria, Actinobacteria, and Bacteroidetes across the range of soil pHs. In addition, soil pH explains a significant portion of the variability associated with observed changes in the phylogenetic structure within each dominant lineage. The overall phylogenetic diversity of the bacterial communities was also correlated with soil pH (R2 = 0.50), with peak diversity in soils with near-neutral pHs. Together, these results suggest that the structure of soil bacterial communities is predictable, to some degree, across larger spatial scales, and the effect of soil pH on bacterial community composition is evident at even relatively coarse levels of taxonomic resolution."	ENVO:tropical soil	TCAG	Pyrosequencing_based_assessment_of_soil_pH_as_a_predictor_of_soil_bacterial_community_structure_at_the_continental_scale	PE7_V2	0	bacterial biogeography	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	19502440	NA	NA	NA	Chris Lauber	CCME	GAZ:Peru	silty clay	lauber_88_soils	0-0.05	NA	NA	88	NA	n	NA	NA	Lauber_88_soils	pyrosequencing	NA	NA	-1600	soil metagenome	NA	ENVO:Tropical humid forests	FLX	PE7_V2	NA	4000	5.51	NA	Lauber88Soils	Pyrosequencing_based_assessment_of_soil_pH_as_a_predictor_of_soil_bacterial_community_structure_at_the_continental_scale	410658	CCME	oxisol	NA	NA	103	19502440	lauber_88_soils	NA	y	NA	PE7_V2	ENVO:soil	16S rRNA	NA	NA	V2	CCME	V2	NA	NA	Engencore	NA	-12.65	NA	NA	soil metagenome
B1.489537	GGACGTCACAGT	YATGCTGCCTCCCGTAGGAGT	jesse.stombaugh@colorado.edu	NA	restroom surface	Bathroom	NA	NA	NA	n	12/13/10	restroom surface metagenome	NA	GSQCOUB	NA	mimarks-survey	NA	NA	NA	NA	0	NA	NA	1624.1	10-Nov	NA	0	NA	NA	NA	NA	NA	NA	NA	NA	NA	NA	"1,filter"	A	NA	Microbial biogeography of public restroom surfaces	NA	NA	-105.27055	y	NA	CCME	NA	NA	NA	"We spend the majority of our lives indoors where we are constantly exposed to bacteria residing on surfaces. However, the diversity of these surface-associated communities is largely unknown. We explored the biogeographical patterns exhibited by bacteria across ten surfaces within each of twelve public restrooms. Using high-throughput barcoded pyrosequencing of the 16S rRNA gene, we identified 19 bacterial phyla across all surfaces. Most sequences belonged to four phyla: Actinobacteria, Bacteriodetes, Firmicutes and Proteobacteria. The communities clustered into three general categories: those found on surfaces associated with toilets, those on the restroom floor, and those found on surfaces routinely touched with hands. On toilet surfaces, gut-associated taxa were more prevalent, suggesting fecal contamination of these surfaces. Floor surfaces were the most diverse of all communities and contained several taxa commonly found in soils. Skin-associated bacteria, especially the Propionibacteriaceae, dominated surfaces routinely touched with our hands. Certain taxa were more common in female than in male restrooms as vagina-associated Lactobacillaceae were widely distributed in female restrooms, likely from urine contamination. Use of the SourceTracker algorithm confirmed many of our taxonomic observations as human skin was the primary source of bacteria on restroom surfaces. Overall, these results demonstrate that restroom surfaces host relatively diverse microbial communities dominated by human-associated bacteria with clear linkages between communities on or in different body sites and those communities found on restroom surfaces. More generally, this work is relevant to the public health field as we show that human associated microbes are commonly found on restroom surfaces suggesting that bacterial pathogens could readily be transmitted between individuals by the touching of surfaces. Furthermore, we demonstrate that we can use high-throughput analyses of bacterial communities to determine sources of bacteria on indoor surfaces, an approach which could be used to track pathogen transmission and test the efficacy of hygiene practices."	ENVO:building	TCAG	Microbial biogeography of public restroom surfaces	NA	0	"Ten surfaces (door handles into and out of the restroom, handles into and out of a restroom stall, faucet handles, soap dispenser, toilet seat, toilet flush handle, floor around the toilet and floor around the sink) in six male and six female restrooms evenly distributed across two buildings on the University of Colorado at Boulder campus were sampled on a single day in November 2010. Surfaces where sampled using sterile, cotton-tipped swabs as described previously?[14],?[15]. As the 12 restrooms were nearly identical in design, we were able to swab the same area at each location between restrooms. In order to characterize tap water communities as a potential source of bacteria, 1 L of faucet water from six of the restrooms (each building having the same water source for each restroom sampled) was collected and filtered through 0.2 ?m bottle top filters (Nalgene, Rochester, NY, USA)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	22132229	NA	NA	NA	Gilberto Flores	CCME	GAZ:United States of America	NA	NA	0	NA	NA	NA	NA	n	NA	Noah Fierer	Flores_restroom_surface_biogeography	pyrosequencing	NA	NA	NA	restroom surface metagenome	1	ENVO:city	Titanium	NA	NA	NA	NA	NA	flores_restroom_surface_biogeography	restroom surface biogeography	256318	CCME	NA	NA	NA	1335	"Genomic DNA was extracted from the swabs and filters using the MO BIO PowerSoil DNA isolation kit following the manufacturer's protocol with the modifications of Fierer?et al.?[14]. A portion of the 16 S rRNA gene spanning the V1_V2 regions was amplified using the primer set (27 F/338R), PCR mixture conditions and thermal cycling conditions described in Fierer?et al.?[15]. PCR amplicons of triplicate reactions for each sample were pooled at approximately equal amounts and pyrosequenced at 454 Life Sciences (Branford, CT, USA) on their GS Junior system. A total of 337,333 high-quality partial 16 S rRNA gene sequences were obtained from 101 of the 120 surface samples collected, averaging approximately 3,340 sequences per sample (ranging from 513_6,771) (Table S1) in 4 GS Junior runs, with the best run containing 116,004 high-quality reads. An additional 16,416 sequences (ranging from 2161_5084 per sample) were generated for five of the six water samples collected for source tracking analysis. Each sample was amplified with a unique barcode to enable multiplexing in the GS Junior runs. The barcoded sequencing reads can be separated by data analysis software providing high confidence in assigning sequencing read to each sample. Sequence data generated as part of this study is available upon request by contacting the corresponding author."	NA	NA	y	NA	NA	ENVO:water	16S rRNA	Ekeley	NA	V2	CCME	NA	Water	NA	NA	NA	40.01499	NA	NA	Bathroom
B2.489526	GTACTCTAGACT	YATGCTGCCTCCCGTAGGAGT	jesse.stombaugh@colorado.edu	NA	restroom surface	Bathroom	NA	NA	NA	n	12/13/10	restroom surface metagenome	NA	GSQCOUB	NA	mimarks-survey	NA	NA	NA	NA	0	NA	NA	1624.1	10-Nov	NA	0	NA	NA	NA	NA	NA	NA	NA	NA	NA	NA	"1,filter"	B	NA	Microbial biogeography of public restroom surfaces	NA	NA	-105.27055	y	NA	CCME	NA	NA	NA	"We spend the majority of our lives indoors where we are constantly exposed to bacteria residing on surfaces. However, the diversity of these surface-associated communities is largely unknown. We explored the biogeographical patterns exhibited by bacteria across ten surfaces within each of twelve public restrooms. Using high-throughput barcoded pyrosequencing of the 16S rRNA gene, we identified 19 bacterial phyla across all surfaces. Most sequences belonged to four phyla: Actinobacteria, Bacteriodetes, Firmicutes and Proteobacteria. The communities clustered into three general categories: those found on surfaces associated with toilets, those on the restroom floor, and those found on surfaces routinely touched with hands. On toilet surfaces, gut-associated taxa were more prevalent, suggesting fecal contamination of these surfaces. Floor surfaces were the most diverse of all communities and contained several taxa commonly found in soils. Skin-associated bacteria, especially the Propionibacteriaceae, dominated surfaces routinely touched with our hands. Certain taxa were more common in female than in male restrooms as vagina-associated Lactobacillaceae were widely distributed in female restrooms, likely from urine contamination. Use of the SourceTracker algorithm confirmed many of our taxonomic observations as human skin was the primary source of bacteria on restroom surfaces. Overall, these results demonstrate that restroom surfaces host relatively diverse microbial communities dominated by human-associated bacteria with clear linkages between communities on or in different body sites and those communities found on restroom surfaces. More generally, this work is relevant to the public health field as we show that human associated microbes are commonly found on restroom surfaces suggesting that bacterial pathogens could readily be transmitted between individuals by the touching of surfaces. Furthermore, we demonstrate that we can use high-throughput analyses of bacterial communities to determine sources of bacteria on indoor surfaces, an approach which could be used to track pathogen transmission and test the efficacy of hygiene practices."	ENVO:building	TCAG	Microbial biogeography of public restroom surfaces	NA	0	"Ten surfaces (door handles into and out of the restroom, handles into and out of a restroom stall, faucet handles, soap dispenser, toilet seat, toilet flush handle, floor around the toilet and floor around the sink) in six male and six female restrooms evenly distributed across two buildings on the University of Colorado at Boulder campus were sampled on a single day in November 2010. Surfaces where sampled using sterile, cotton-tipped swabs as described previously?[14],?[15]. As the 12 restrooms were nearly identical in design, we were able to swab the same area at each location between restrooms. In order to characterize tap water communities as a potential source of bacteria, 1 L of faucet water from six of the restrooms (each building having the same water source for each restroom sampled) was collected and filtered through 0.2 ?m bottle top filters (Nalgene, Rochester, NY, USA)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	22132229	NA	NA	NA	Gilberto Flores	CCME	GAZ:United States of America	NA	NA	0	NA	NA	NA	NA	n	NA	Noah Fierer	Flores_restroom_surface_biogeography	pyrosequencing	NA	NA	NA	restroom surface metagenome	2	ENVO:city	Titanium	NA	NA	NA	NA	NA	flores_restroom_surface_biogeography	restroom surface biogeography	256318	CCME	NA	NA	NA	1335	"Genomic DNA was extracted from the swabs and filters using the MO BIO PowerSoil DNA isolation kit following the manufacturer's protocol with the modifications of Fierer?et al.?[14]. A portion of the 16 S rRNA gene spanning the V1_V2 regions was amplified using the primer set (27 F/338R), PCR mixture conditions and thermal cycling conditions described in Fierer?et al.?[15]. PCR amplicons of triplicate reactions for each sample were pooled at approximately equal amounts and pyrosequenced at 454 Life Sciences (Branford, CT, USA) on their GS Junior system. A total of 337,333 high-quality partial 16 S rRNA gene sequences were obtained from 101 of the 120 surface samples collected, averaging approximately 3,340 sequences per sample (ranging from 513_6,771) (Table S1) in 4 GS Junior runs, with the best run containing 116,004 high-quality reads. An additional 16,416 sequences (ranging from 2161_5084 per sample) were generated for five of the six water samples collected for source tracking analysis. Each sample was amplified with a unique barcode to enable multiplexing in the GS Junior runs. The barcoded sequencing reads can be separated by data analysis software providing high confidence in assigning sequencing read to each sample. Sequence data generated as part of this study is available upon request by contacting the corresponding author."	NA	NA	y	NA	NA	ENVO:water	16S rRNA	Ekeley	NA	V2	CCME	NA	Water	NA	NA	NA	40.01499	NA	NA	Bathroom
B3.489528	GTATGTTGCTCA	YATGCTGCCTCCCGTAGGAGT	jesse.stombaugh@colorado.edu	NA	restroom surface	Bathroom	NA	NA	NA	n	12/13/10	restroom surface metagenome	NA	GSQCOUB	NA	mimarks-survey	NA	NA	NA	NA	0	NA	NA	1624.1	10-Nov	NA	0	NA	NA	NA	NA	NA	NA	NA	NA	NA	NA	"1,filter"	C	NA	Microbial biogeography of public restroom surfaces	NA	NA	-105.27055	y	NA	CCME	NA	NA	NA	"We spend the majority of our lives indoors where we are constantly exposed to bacteria residing on surfaces. However, the diversity of these surface-associated communities is largely unknown. We explored the biogeographical patterns exhibited by bacteria across ten surfaces within each of twelve public restrooms. Using high-throughput barcoded pyrosequencing of the 16S rRNA gene, we identified 19 bacterial phyla across all surfaces. Most sequences belonged to four phyla: Actinobacteria, Bacteriodetes, Firmicutes and Proteobacteria. The communities clustered into three general categories: those found on surfaces associated with toilets, those on the restroom floor, and those found on surfaces routinely touched with hands. On toilet surfaces, gut-associated taxa were more prevalent, suggesting fecal contamination of these surfaces. Floor surfaces were the most diverse of all communities and contained several taxa commonly found in soils. Skin-associated bacteria, especially the Propionibacteriaceae, dominated surfaces routinely touched with our hands. Certain taxa were more common in female than in male restrooms as vagina-associated Lactobacillaceae were widely distributed in female restrooms, likely from urine contamination. Use of the SourceTracker algorithm confirmed many of our taxonomic observations as human skin was the primary source of bacteria on restroom surfaces. Overall, these results demonstrate that restroom surfaces host relatively diverse microbial communities dominated by human-associated bacteria with clear linkages between communities on or in different body sites and those communities found on restroom surfaces. More generally, this work is relevant to the public health field as we show that human associated microbes are commonly found on restroom surfaces suggesting that bacterial pathogens could readily be transmitted between individuals by the touching of surfaces. Furthermore, we demonstrate that we can use high-throughput analyses of bacterial communities to determine sources of bacteria on indoor surfaces, an approach which could be used to track pathogen transmission and test the efficacy of hygiene practices."	ENVO:building	TCAG	Microbial biogeography of public restroom surfaces	NA	0	"Ten surfaces (door handles into and out of the restroom, handles into and out of a restroom stall, faucet handles, soap dispenser, toilet seat, toilet flush handle, floor around the toilet and floor around the sink) in six male and six female restrooms evenly distributed across two buildings on the University of Colorado at Boulder campus were sampled on a single day in November 2010. Surfaces where sampled using sterile, cotton-tipped swabs as described previously?[14],?[15]. As the 12 restrooms were nearly identical in design, we were able to swab the same area at each location between restrooms. In order to characterize tap water communities as a potential source of bacteria, 1 L of faucet water from six of the restrooms (each building having the same water source for each restroom sampled) was collected and filtered through 0.2 ?m bottle top filters (Nalgene, Rochester, NY, USA)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	22132229	NA	NA	NA	Gilberto Flores	CCME	GAZ:United States of America	NA	NA	0	NA	NA	NA	NA	n	NA	Noah Fierer	Flores_restroom_surface_biogeography	pyrosequencing	NA	NA	NA	restroom surface metagenome	3	ENVO:city	Titanium	NA	NA	NA	NA	NA	flores_restroom_surface_biogeography	restroom surface biogeography	256318	CCME	NA	NA	NA	1335	"Genomic DNA was extracted from the swabs and filters using the MO BIO PowerSoil DNA isolation kit following the manufacturer's protocol with the modifications of Fierer?et al.?[14]. A portion of the 16 S rRNA gene spanning the V1_V2 regions was amplified using the primer set (27 F/338R), PCR mixture conditions and thermal cycling conditions described in Fierer?et al.?[15]. PCR amplicons of triplicate reactions for each sample were pooled at approximately equal amounts and pyrosequenced at 454 Life Sciences (Branford, CT, USA) on their GS Junior system. A total of 337,333 high-quality partial 16 S rRNA gene sequences were obtained from 101 of the 120 surface samples collected, averaging approximately 3,340 sequences per sample (ranging from 513_6,771) (Table S1) in 4 GS Junior runs, with the best run containing 116,004 high-quality reads. An additional 16,416 sequences (ranging from 2161_5084 per sample) were generated for five of the six water samples collected for source tracking analysis. Each sample was amplified with a unique barcode to enable multiplexing in the GS Junior runs. The barcoded sequencing reads can be separated by data analysis software providing high confidence in assigning sequencing read to each sample. Sequence data generated as part of this study is available upon request by contacting the corresponding author."	NA	NA	y	NA	NA	ENVO:water	16S rRNA	Ekeley	NA	V2	CCME	NA	Water	NA	NA	NA	40.01499	NA	NA	Bathroom
B5.489455	GTGTACCTATCA	YATGCTGCCTCCCGTAGGAGT	jesse.stombaugh@colorado.edu	NA	restroom surface	Bathroom	NA	NA	NA	n	12/13/10	restroom surface metagenome	NA	GSQCOUB	NA	mimarks-survey	NA	NA	NA	NA	0	NA	NA	1624.1	10-Nov	NA	0	NA	NA	NA	NA	NA	NA	NA	NA	NA	NA	"1,filter"	B	NA	Microbial biogeography of public restroom surfaces	NA	NA	-105.27055	y	NA	CCME	NA	NA	NA	"We spend the majority of our lives indoors where we are constantly exposed to bacteria residing on surfaces. However, the diversity of these surface-associated communities is largely unknown. We explored the biogeographical patterns exhibited by bacteria across ten surfaces within each of twelve public restrooms. Using high-throughput barcoded pyrosequencing of the 16S rRNA gene, we identified 19 bacterial phyla across all surfaces. Most sequences belonged to four phyla: Actinobacteria, Bacteriodetes, Firmicutes and Proteobacteria. The communities clustered into three general categories: those found on surfaces associated with toilets, those on the restroom floor, and those found on surfaces routinely touched with hands. On toilet surfaces, gut-associated taxa were more prevalent, suggesting fecal contamination of these surfaces. Floor surfaces were the most diverse of all communities and contained several taxa commonly found in soils. Skin-associated bacteria, especially the Propionibacteriaceae, dominated surfaces routinely touched with our hands. Certain taxa were more common in female than in male restrooms as vagina-associated Lactobacillaceae were widely distributed in female restrooms, likely from urine contamination. Use of the SourceTracker algorithm confirmed many of our taxonomic observations as human skin was the primary source of bacteria on restroom surfaces. Overall, these results demonstrate that restroom surfaces host relatively diverse microbial communities dominated by human-associated bacteria with clear linkages between communities on or in different body sites and those communities found on restroom surfaces. More generally, this work is relevant to the public health field as we show that human associated microbes are commonly found on restroom surfaces suggesting that bacterial pathogens could readily be transmitted between individuals by the touching of surfaces. Furthermore, we demonstrate that we can use high-throughput analyses of bacterial communities to determine sources of bacteria on indoor surfaces, an approach which could be used to track pathogen transmission and test the efficacy of hygiene practices."	ENVO:building	TCAG	Microbial biogeography of public restroom surfaces	NA	0	"Ten surfaces (door handles into and out of the restroom, handles into and out of a restroom stall, faucet handles, soap dispenser, toilet seat, toilet flush handle, floor around the toilet and floor around the sink) in six male and six female restrooms evenly distributed across two buildings on the University of Colorado at Boulder campus were sampled on a single day in November 2010. Surfaces where sampled using sterile, cotton-tipped swabs as described previously?[14],?[15]. As the 12 restrooms were nearly identical in design, we were able to swab the same area at each location between restrooms. In order to characterize tap water communities as a potential source of bacteria, 1 L of faucet water from six of the restrooms (each building having the same water source for each restroom sampled) was collected and filtered through 0.2 ?m bottle top filters (Nalgene, Rochester, NY, USA)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	22132229	NA	NA	NA	Gilberto Flores	CCME	GAZ:United States of America	NA	NA	0	NA	NA	NA	NA	n	NA	Noah Fierer	Flores_restroom_surface_biogeography	pyrosequencing	NA	NA	NA	restroom surface metagenome	2	ENVO:city	Titanium	NA	NA	NA	NA	NA	flores_restroom_surface_biogeography	restroom surface biogeography	256318	CCME	NA	NA	NA	1335	"Genomic DNA was extracted from the swabs and filters using the MO BIO PowerSoil DNA isolation kit following the manufacturer's protocol with the modifications of Fierer?et al.?[14]. A portion of the 16 S rRNA gene spanning the V1_V2 regions was amplified using the primer set (27 F/338R), PCR mixture conditions and thermal cycling conditions described in Fierer?et al.?[15]. PCR amplicons of triplicate reactions for each sample were pooled at approximately equal amounts and pyrosequenced at 454 Life Sciences (Branford, CT, USA) on their GS Junior system. A total of 337,333 high-quality partial 16 S rRNA gene sequences were obtained from 101 of the 120 surface samples collected, averaging approximately 3,340 sequences per sample (ranging from 513_6,771) (Table S1) in 4 GS Junior runs, with the best run containing 116,004 high-quality reads. An additional 16,416 sequences (ranging from 2161_5084 per sample) were generated for five of the six water samples collected for source tracking analysis. Each sample was amplified with a unique barcode to enable multiplexing in the GS Junior runs. The barcoded sequencing reads can be separated by data analysis software providing high confidence in assigning sequencing read to each sample. Sequence data generated as part of this study is available upon request by contacting the corresponding author."	NA	NA	y	NA	NA	ENVO:water	16S rRNA	Porter	NA	V2	CCME	NA	Water	NA	NA	NA	40.01499	NA	NA	Bathroom
B6.489449	TACACACATGGC	YATGCTGCCTCCCGTAGGAGT	jesse.stombaugh@colorado.edu	NA	restroom surface	Bathroom	NA	NA	NA	n	12/13/10	restroom surface metagenome	NA	GSQCOUB	NA	mimarks-survey	NA	NA	NA	NA	0	NA	NA	1624.1	10-Nov	NA	0	NA	NA	NA	NA	NA	NA	NA	NA	NA	NA	"1,filter"	C	NA	Microbial biogeography of public restroom surfaces	NA	NA	-105.27055	y	NA	CCME	NA	NA	NA	"We spend the majority of our lives indoors where we are constantly exposed to bacteria residing on surfaces. However, the diversity of these surface-associated communities is largely unknown. We explored the biogeographical patterns exhibited by bacteria across ten surfaces within each of twelve public restrooms. Using high-throughput barcoded pyrosequencing of the 16S rRNA gene, we identified 19 bacterial phyla across all surfaces. Most sequences belonged to four phyla: Actinobacteria, Bacteriodetes, Firmicutes and Proteobacteria. The communities clustered into three general categories: those found on surfaces associated with toilets, those on the restroom floor, and those found on surfaces routinely touched with hands. On toilet surfaces, gut-associated taxa were more prevalent, suggesting fecal contamination of these surfaces. Floor surfaces were the most diverse of all communities and contained several taxa commonly found in soils. Skin-associated bacteria, especially the Propionibacteriaceae, dominated surfaces routinely touched with our hands. Certain taxa were more common in female than in male restrooms as vagina-associated Lactobacillaceae were widely distributed in female restrooms, likely from urine contamination. Use of the SourceTracker algorithm confirmed many of our taxonomic observations as human skin was the primary source of bacteria on restroom surfaces. Overall, these results demonstrate that restroom surfaces host relatively diverse microbial communities dominated by human-associated bacteria with clear linkages between communities on or in different body sites and those communities found on restroom surfaces. More generally, this work is relevant to the public health field as we show that human associated microbes are commonly found on restroom surfaces suggesting that bacterial pathogens could readily be transmitted between individuals by the touching of surfaces. Furthermore, we demonstrate that we can use high-throughput analyses of bacterial communities to determine sources of bacteria on indoor surfaces, an approach which could be used to track pathogen transmission and test the efficacy of hygiene practices."	ENVO:building	TCAG	Microbial biogeography of public restroom surfaces	NA	0	"Ten surfaces (door handles into and out of the restroom, handles into and out of a restroom stall, faucet handles, soap dispenser, toilet seat, toilet flush handle, floor around the toilet and floor around the sink) in six male and six female restrooms evenly distributed across two buildings on the University of Colorado at Boulder campus were sampled on a single day in November 2010. Surfaces where sampled using sterile, cotton-tipped swabs as described previously?[14],?[15]. As the 12 restrooms were nearly identical in design, we were able to swab the same area at each location between restrooms. In order to characterize tap water communities as a potential source of bacteria, 1 L of faucet water from six of the restrooms (each building having the same water source for each restroom sampled) was collected and filtered through 0.2 ?m bottle top filters (Nalgene, Rochester, NY, USA)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	22132229	NA	NA	NA	Gilberto Flores	CCME	GAZ:United States of America	NA	NA	0	NA	NA	NA	NA	n	NA	Noah Fierer	Flores_restroom_surface_biogeography	pyrosequencing	NA	NA	NA	restroom surface metagenome	3	ENVO:city	Titanium	NA	NA	NA	NA	NA	flores_restroom_surface_biogeography	restroom surface biogeography	256318	CCME	NA	NA	NA	1335	"Genomic DNA was extracted from the swabs and filters using the MO BIO PowerSoil DNA isolation kit following the manufacturer's protocol with the modifications of Fierer?et al.?[14]. A portion of the 16 S rRNA gene spanning the V1_V2 regions was amplified using the primer set (27 F/338R), PCR mixture conditions and thermal cycling conditions described in Fierer?et al.?[15]. PCR amplicons of triplicate reactions for each sample were pooled at approximately equal amounts and pyrosequenced at 454 Life Sciences (Branford, CT, USA) on their GS Junior system. A total of 337,333 high-quality partial 16 S rRNA gene sequences were obtained from 101 of the 120 surface samples collected, averaging approximately 3,340 sequences per sample (ranging from 513_6,771) (Table S1) in 4 GS Junior runs, with the best run containing 116,004 high-quality reads. An additional 16,416 sequences (ranging from 2161_5084 per sample) were generated for five of the six water samples collected for source tracking analysis. Each sample was amplified with a unique barcode to enable multiplexing in the GS Junior runs. The barcoded sequencing reads can be separated by data analysis software providing high confidence in assigning sequencing read to each sample. Sequence data generated as part of this study is available upon request by contacting the corresponding author."	NA	NA	y	NA	NA	ENVO:water	16S rRNA	Porter	NA	V2	CCME	NA	Water	NA	NA	NA	40.01499	NA	NA	Bathroom
EKAF1.489470	GACACTCGAATC	YATGCTGCCTCCCGTAGGAGT	jesse.stombaugh@colorado.edu	NA	restroom surface	Door_handle_area_into_restroom	NA	NA	NA	n	12/13/10	restroom surface metagenome	NA	GFL0IUQ	NA	mimarks-survey	NA	NA	NA	NA	0	NA	NA	1624.1	10-Nov	Female	0	NA	NA	NA	NA	NA	NA	NA	NA	NA	NA	"1,swab"	A	NA	Microbial biogeography of public restroom surfaces	NA	NA	-105.27055	y	NA	CCME	NA	NA	NA	"We spend the majority of our lives indoors where we are constantly exposed to bacteria residing on surfaces. However, the diversity of these surface-associated communities is largely unknown. We explored the biogeographical patterns exhibited by bacteria across ten surfaces within each of twelve public restrooms. Using high-throughput barcoded pyrosequencing of the 16S rRNA gene, we identified 19 bacterial phyla across all surfaces. Most sequences belonged to four phyla: Actinobacteria, Bacteriodetes, Firmicutes and Proteobacteria. The communities clustered into three general categories: those found on surfaces associated with toilets, those on the restroom floor, and those found on surfaces routinely touched with hands. On toilet surfaces, gut-associated taxa were more prevalent, suggesting fecal contamination of these surfaces. Floor surfaces were the most diverse of all communities and contained several taxa commonly found in soils. Skin-associated bacteria, especially the Propionibacteriaceae, dominated surfaces routinely touched with our hands. Certain taxa were more common in female than in male restrooms as vagina-associated Lactobacillaceae were widely distributed in female restrooms, likely from urine contamination. Use of the SourceTracker algorithm confirmed many of our taxonomic observations as human skin was the primary source of bacteria on restroom surfaces. Overall, these results demonstrate that restroom surfaces host relatively diverse microbial communities dominated by human-associated bacteria with clear linkages between communities on or in different body sites and those communities found on restroom surfaces. More generally, this work is relevant to the public health field as we show that human associated microbes are commonly found on restroom surfaces suggesting that bacterial pathogens could readily be transmitted between individuals by the touching of surfaces. Furthermore, we demonstrate that we can use high-throughput analyses of bacterial communities to determine sources of bacteria on indoor surfaces, an approach which could be used to track pathogen transmission and test the efficacy of hygiene practices."	ENVO:building	TCAG	Microbial biogeography of public restroom surfaces	NA	0	"Ten surfaces (door handles into and out of the restroom, handles into and out of a restroom stall, faucet handles, soap dispenser, toilet seat, toilet flush handle, floor around the toilet and floor around the sink) in six male and six female restrooms evenly distributed across two buildings on the University of Colorado at Boulder campus were sampled on a single day in November 2010. Surfaces where sampled using sterile, cotton-tipped swabs as described previously?[14],?[15]. As the 12 restrooms were nearly identical in design, we were able to swab the same area at each location between restrooms. In order to characterize tap water communities as a potential source of bacteria, 1 L of faucet water from six of the restrooms (each building having the same water source for each restroom sampled) was collected and filtered through 0.2 ?m bottle top filters (Nalgene, Rochester, NY, USA)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	22132229	NA	NA	NA	Gilberto Flores	CCME	GAZ:United States of America	NA	NA	0	NA	NA	NA	NA	n	NA	Noah Fierer	Flores_restroom_surface_biogeography	pyrosequencing	NA	NA	NA	restroom surface metagenome	1	ENVO:city	Titanium	NA	NA	NA	NA	NA	flores_restroom_surface_biogeography	restroom surface biogeography	256318	CCME	NA	NA	NA	1335	"Genomic DNA was extracted from the swabs and filters using the MO BIO PowerSoil DNA isolation kit following the manufacturer's protocol with the modifications of Fierer?et al.?[14]. A portion of the 16 S rRNA gene spanning the V1_V2 regions was amplified using the primer set (27 F/338R), PCR mixture conditions and thermal cycling conditions described in Fierer?et al.?[15]. PCR amplicons of triplicate reactions for each sample were pooled at approximately equal amounts and pyrosequenced at 454 Life Sciences (Branford, CT, USA) on their GS Junior system. A total of 337,333 high-quality partial 16 S rRNA gene sequences were obtained from 101 of the 120 surface samples collected, averaging approximately 3,340 sequences per sample (ranging from 513_6,771) (Table S1) in 4 GS Junior runs, with the best run containing 116,004 high-quality reads. An additional 16,416 sequences (ranging from 2161_5084 per sample) were generated for five of the six water samples collected for source tracking analysis. Each sample was amplified with a unique barcode to enable multiplexing in the GS Junior runs. The barcoded sequencing reads can be separated by data analysis software providing high confidence in assigning sequencing read to each sample. Sequence data generated as part of this study is available upon request by contacting the corresponding author."	NA	NA	y	NA	NA	ENVO:surface	16S rRNA	Ekeley	NA	V2	CCME	NA	Door in	NA	NA	NA	40.01499	NA	NA	Door handle area into restroom
EKAM1.489459	CTGTTCGTAGAG	YATGCTGCCTCCCGTAGGAGT	jesse.stombaugh@colorado.edu	NA	restroom surface	Door_handle_area_into_restroom	NA	NA	NA	n	12/13/10	restroom surface metagenome	NA	GFL0IUQ	NA	mimarks-survey	NA	NA	NA	NA	0	NA	NA	1624.1	10-Nov	Male	0	NA	NA	NA	NA	NA	NA	NA	NA	NA	NA	"1,swab"	A	NA	Microbial biogeography of public restroom surfaces	NA	NA	-105.27055	y	NA	CCME	NA	NA	NA	"We spend the majority of our lives indoors where we are constantly exposed to bacteria residing on surfaces. However, the diversity of these surface-associated communities is largely unknown. We explored the biogeographical patterns exhibited by bacteria across ten surfaces within each of twelve public restrooms. Using high-throughput barcoded pyrosequencing of the 16S rRNA gene, we identified 19 bacterial phyla across all surfaces. Most sequences belonged to four phyla: Actinobacteria, Bacteriodetes, Firmicutes and Proteobacteria. The communities clustered into three general categories: those found on surfaces associated with toilets, those on the restroom floor, and those found on surfaces routinely touched with hands. On toilet surfaces, gut-associated taxa were more prevalent, suggesting fecal contamination of these surfaces. Floor surfaces were the most diverse of all communities and contained several taxa commonly found in soils. Skin-associated bacteria, especially the Propionibacteriaceae, dominated surfaces routinely touched with our hands. Certain taxa were more common in female than in male restrooms as vagina-associated Lactobacillaceae were widely distributed in female restrooms, likely from urine contamination. Use of the SourceTracker algorithm confirmed many of our taxonomic observations as human skin was the primary source of bacteria on restroom surfaces. Overall, these results demonstrate that restroom surfaces host relatively diverse microbial communities dominated by human-associated bacteria with clear linkages between communities on or in different body sites and those communities found on restroom surfaces. More generally, this work is relevant to the public health field as we show that human associated microbes are commonly found on restroom surfaces suggesting that bacterial pathogens could readily be transmitted between individuals by the touching of surfaces. Furthermore, we demonstrate that we can use high-throughput analyses of bacterial communities to determine sources of bacteria on indoor surfaces, an approach which could be used to track pathogen transmission and test the efficacy of hygiene practices."	ENVO:building	TCAG	Microbial biogeography of public restroom surfaces	NA	0	"Ten surfaces (door handles into and out of the restroom, handles into and out of a restroom stall, faucet handles, soap dispenser, toilet seat, toilet flush handle, floor around the toilet and floor around the sink) in six male and six female restrooms evenly distributed across two buildings on the University of Colorado at Boulder campus were sampled on a single day in November 2010. Surfaces where sampled using sterile, cotton-tipped swabs as described previously?[14],?[15]. As the 12 restrooms were nearly identical in design, we were able to swab the same area at each location between restrooms. In order to characterize tap water communities as a potential source of bacteria, 1 L of faucet water from six of the restrooms (each building having the same water source for each restroom sampled) was collected and filtered through 0.2 ?m bottle top filters (Nalgene, Rochester, NY, USA)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	22132229	NA	NA	NA	Gilberto Flores	CCME	GAZ:United States of America	NA	NA	0	NA	NA	NA	NA	n	NA	Noah Fierer	Flores_restroom_surface_biogeography	pyrosequencing	NA	NA	NA	restroom surface metagenome	1	ENVO:city	Titanium	NA	NA	NA	NA	NA	flores_restroom_surface_biogeography	restroom surface biogeography	256318	CCME	NA	NA	NA	1335	"Genomic DNA was extracted from the swabs and filters using the MO BIO PowerSoil DNA isolation kit following the manufacturer's protocol with the modifications of Fierer?et al.?[14]. A portion of the 16 S rRNA gene spanning the V1_V2 regions was amplified using the primer set (27 F/338R), PCR mixture conditions and thermal cycling conditions described in Fierer?et al.?[15]. PCR amplicons of triplicate reactions for each sample were pooled at approximately equal amounts and pyrosequenced at 454 Life Sciences (Branford, CT, USA) on their GS Junior system. A total of 337,333 high-quality partial 16 S rRNA gene sequences were obtained from 101 of the 120 surface samples collected, averaging approximately 3,340 sequences per sample (ranging from 513_6,771) (Table S1) in 4 GS Junior runs, with the best run containing 116,004 high-quality reads. An additional 16,416 sequences (ranging from 2161_5084 per sample) were generated for five of the six water samples collected for source tracking analysis. Each sample was amplified with a unique barcode to enable multiplexing in the GS Junior runs. The barcoded sequencing reads can be separated by data analysis software providing high confidence in assigning sequencing read to each sample. Sequence data generated as part of this study is available upon request by contacting the corresponding author."	NA	NA	y	NA	NA	ENVO:surface	16S rRNA	Ekeley	NA	V2	CCME	NA	Door in	NA	NA	NA	40.01499	NA	NA	Door handle area into restroom
EKBF1.489501	GAGAATACGTGA	YATGCTGCCTCCCGTAGGAGT	jesse.stombaugh@colorado.edu	NA	restroom surface	Door_handle_area_into_restroom	NA	NA	NA	n	12/13/10	restroom surface metagenome	NA	GFL0IUQ	NA	mimarks-survey	NA	NA	NA	NA	0	NA	NA	1624.1	10-Nov	Female	0	NA	NA	NA	NA	NA	NA	NA	NA	NA	NA	"1,swab"	B	NA	Microbial biogeography of public restroom surfaces	NA	NA	-105.27055	y	NA	CCME	NA	NA	NA	"We spend the majority of our lives indoors where we are constantly exposed to bacteria residing on surfaces. However, the diversity of these surface-associated communities is largely unknown. We explored the biogeographical patterns exhibited by bacteria across ten surfaces within each of twelve public restrooms. Using high-throughput barcoded pyrosequencing of the 16S rRNA gene, we identified 19 bacterial phyla across all surfaces. Most sequences belonged to four phyla: Actinobacteria, Bacteriodetes, Firmicutes and Proteobacteria. The communities clustered into three general categories: those found on surfaces associated with toilets, those on the restroom floor, and those found on surfaces routinely touched with hands. On toilet surfaces, gut-associated taxa were more prevalent, suggesting fecal contamination of these surfaces. Floor surfaces were the most diverse of all communities and contained several taxa commonly found in soils. Skin-associated bacteria, especially the Propionibacteriaceae, dominated surfaces routinely touched with our hands. Certain taxa were more common in female than in male restrooms as vagina-associated Lactobacillaceae were widely distributed in female restrooms, likely from urine contamination. Use of the SourceTracker algorithm confirmed many of our taxonomic observations as human skin was the primary source of bacteria on restroom surfaces. Overall, these results demonstrate that restroom surfaces host relatively diverse microbial communities dominated by human-associated bacteria with clear linkages between communities on or in different body sites and those communities found on restroom surfaces. More generally, this work is relevant to the public health field as we show that human associated microbes are commonly found on restroom surfaces suggesting that bacterial pathogens could readily be transmitted between individuals by the touching of surfaces. Furthermore, we demonstrate that we can use high-throughput analyses of bacterial communities to determine sources of bacteria on indoor surfaces, an approach which could be used to track pathogen transmission and test the efficacy of hygiene practices."	ENVO:building	TCAG	Microbial biogeography of public restroom surfaces	NA	0	"Ten surfaces (door handles into and out of the restroom, handles into and out of a restroom stall, faucet handles, soap dispenser, toilet seat, toilet flush handle, floor around the toilet and floor around the sink) in six male and six female restrooms evenly distributed across two buildings on the University of Colorado at Boulder campus were sampled on a single day in November 2010. Surfaces where sampled using sterile, cotton-tipped swabs as described previously?[14],?[15]. As the 12 restrooms were nearly identical in design, we were able to swab the same area at each location between restrooms. In order to characterize tap water communities as a potential source of bacteria, 1 L of faucet water from six of the restrooms (each building having the same water source for each restroom sampled) was collected and filtered through 0.2 ?m bottle top filters (Nalgene, Rochester, NY, USA)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	22132229	NA	NA	NA	Gilberto Flores	CCME	GAZ:United States of America	NA	NA	0	NA	NA	NA	NA	n	NA	Noah Fierer	Flores_restroom_surface_biogeography	pyrosequencing	NA	NA	NA	restroom surface metagenome	2	ENVO:city	Titanium	NA	NA	NA	NA	NA	flores_restroom_surface_biogeography	restroom surface biogeography	256318	CCME	NA	NA	NA	1335	"Genomic DNA was extracted from the swabs and filters using the MO BIO PowerSoil DNA isolation kit following the manufacturer's protocol with the modifications of Fierer?et al.?[14]. A portion of the 16 S rRNA gene spanning the V1_V2 regions was amplified using the primer set (27 F/338R), PCR mixture conditions and thermal cycling conditions described in Fierer?et al.?[15]. PCR amplicons of triplicate reactions for each sample were pooled at approximately equal amounts and pyrosequenced at 454 Life Sciences (Branford, CT, USA) on their GS Junior system. A total of 337,333 high-quality partial 16 S rRNA gene sequences were obtained from 101 of the 120 surface samples collected, averaging approximately 3,340 sequences per sample (ranging from 513_6,771) (Table S1) in 4 GS Junior runs, with the best run containing 116,004 high-quality reads. An additional 16,416 sequences (ranging from 2161_5084 per sample) were generated for five of the six water samples collected for source tracking analysis. Each sample was amplified with a unique barcode to enable multiplexing in the GS Junior runs. The barcoded sequencing reads can be separated by data analysis software providing high confidence in assigning sequencing read to each sample. Sequence data generated as part of this study is available upon request by contacting the corresponding author."	NA	NA	y	NA	NA	ENVO:surface	16S rRNA	Ekeley	NA	V2	CCME	NA	Door in	NA	NA	NA	40.01499	NA	NA	Door handle area into restroom
EKCF1.489542	GATCGCAGGTGT	YATGCTGCCTCCCGTAGGAGT	jesse.stombaugh@colorado.edu	NA	restroom surface	Door_handle_area_into_restroom	NA	NA	NA	n	12/13/10	restroom surface metagenome	NA	GFL0IUQ	NA	mimarks-survey	NA	NA	NA	NA	0	NA	NA	1624.1	10-Nov	Female	0	NA	NA	NA	NA	NA	NA	NA	NA	NA	NA	"1,swab"	C	NA	Microbial biogeography of public restroom surfaces	NA	NA	-105.27055	y	NA	CCME	NA	NA	NA	"We spend the majority of our lives indoors where we are constantly exposed to bacteria residing on surfaces. However, the diversity of these surface-associated communities is largely unknown. We explored the biogeographical patterns exhibited by bacteria across ten surfaces within each of twelve public restrooms. Using high-throughput barcoded pyrosequencing of the 16S rRNA gene, we identified 19 bacterial phyla across all surfaces. Most sequences belonged to four phyla: Actinobacteria, Bacteriodetes, Firmicutes and Proteobacteria. The communities clustered into three general categories: those found on surfaces associated with toilets, those on the restroom floor, and those found on surfaces routinely touched with hands. On toilet surfaces, gut-associated taxa were more prevalent, suggesting fecal contamination of these surfaces. Floor surfaces were the most diverse of all communities and contained several taxa commonly found in soils. Skin-associated bacteria, especially the Propionibacteriaceae, dominated surfaces routinely touched with our hands. Certain taxa were more common in female than in male restrooms as vagina-associated Lactobacillaceae were widely distributed in female restrooms, likely from urine contamination. Use of the SourceTracker algorithm confirmed many of our taxonomic observations as human skin was the primary source of bacteria on restroom surfaces. Overall, these results demonstrate that restroom surfaces host relatively diverse microbial communities dominated by human-associated bacteria with clear linkages between communities on or in different body sites and those communities found on restroom surfaces. More generally, this work is relevant to the public health field as we show that human associated microbes are commonly found on restroom surfaces suggesting that bacterial pathogens could readily be transmitted between individuals by the touching of surfaces. Furthermore, we demonstrate that we can use high-throughput analyses of bacterial communities to determine sources of bacteria on indoor surfaces, an approach which could be used to track pathogen transmission and test the efficacy of hygiene practices."	ENVO:building	TCAG	Microbial biogeography of public restroom surfaces	NA	0	"Ten surfaces (door handles into and out of the restroom, handles into and out of a restroom stall, faucet handles, soap dispenser, toilet seat, toilet flush handle, floor around the toilet and floor around the sink) in six male and six female restrooms evenly distributed across two buildings on the University of Colorado at Boulder campus were sampled on a single day in November 2010. Surfaces where sampled using sterile, cotton-tipped swabs as described previously?[14],?[15]. As the 12 restrooms were nearly identical in design, we were able to swab the same area at each location between restrooms. In order to characterize tap water communities as a potential source of bacteria, 1 L of faucet water from six of the restrooms (each building having the same water source for each restroom sampled) was collected and filtered through 0.2 ?m bottle top filters (Nalgene, Rochester, NY, USA)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	22132229	NA	NA	NA	Gilberto Flores	CCME	GAZ:United States of America	NA	NA	0	NA	NA	NA	NA	n	NA	Noah Fierer	Flores_restroom_surface_biogeography	pyrosequencing	NA	NA	NA	restroom surface metagenome	3	ENVO:city	Titanium	NA	NA	NA	NA	NA	flores_restroom_surface_biogeography	restroom surface biogeography	256318	CCME	NA	NA	NA	1335	"Genomic DNA was extracted from the swabs and filters using the MO BIO PowerSoil DNA isolation kit following the manufacturer's protocol with the modifications of Fierer?et al.?[14]. A portion of the 16 S rRNA gene spanning the V1_V2 regions was amplified using the primer set (27 F/338R), PCR mixture conditions and thermal cycling conditions described in Fierer?et al.?[15]. PCR amplicons of triplicate reactions for each sample were pooled at approximately equal amounts and pyrosequenced at 454 Life Sciences (Branford, CT, USA) on their GS Junior system. A total of 337,333 high-quality partial 16 S rRNA gene sequences were obtained from 101 of the 120 surface samples collected, averaging approximately 3,340 sequences per sample (ranging from 513_6,771) (Table S1) in 4 GS Junior runs, with the best run containing 116,004 high-quality reads. An additional 16,416 sequences (ranging from 2161_5084 per sample) were generated for five of the six water samples collected for source tracking analysis. Each sample was amplified with a unique barcode to enable multiplexing in the GS Junior runs. The barcoded sequencing reads can be separated by data analysis software providing high confidence in assigning sequencing read to each sample. Sequence data generated as part of this study is available upon request by contacting the corresponding author."	NA	NA	y	NA	NA	ENVO:surface	16S rRNA	Ekeley	NA	V2	CCME	NA	Door in	NA	NA	NA	40.01499	NA	NA	Door handle area into restroom
EKCM1.489478	GAGTATGCAGCC	YATGCTGCCTCCCGTAGGAGT	jesse.stombaugh@colorado.edu	NA	restroom surface	Door_handle_area_into_restroom	NA	NA	NA	n	12/13/10	restroom surface metagenome	NA	GFL0IUQ	NA	mimarks-survey	NA	NA	NA	NA	0	NA	NA	1624.1	10-Nov	Male	0	NA	NA	NA	NA	NA	NA	NA	NA	NA	NA	"1,swab"	C	NA	Microbial biogeography of public restroom surfaces	NA	NA	-105.27055	y	NA	CCME	NA	NA	NA	"We spend the majority of our lives indoors where we are constantly exposed to bacteria residing on surfaces. However, the diversity of these surface-associated communities is largely unknown. We explored the biogeographical patterns exhibited by bacteria across ten surfaces within each of twelve public restrooms. Using high-throughput barcoded pyrosequencing of the 16S rRNA gene, we identified 19 bacterial phyla across all surfaces. Most sequences belonged to four phyla: Actinobacteria, Bacteriodetes, Firmicutes and Proteobacteria. The communities clustered into three general categories: those found on surfaces associated with toilets, those on the restroom floor, and those found on surfaces routinely touched with hands. On toilet surfaces, gut-associated taxa were more prevalent, suggesting fecal contamination of these surfaces. Floor surfaces were the most diverse of all communities and contained several taxa commonly found in soils. Skin-associated bacteria, especially the Propionibacteriaceae, dominated surfaces routinely touched with our hands. Certain taxa were more common in female than in male restrooms as vagina-associated Lactobacillaceae were widely distributed in female restrooms, likely from urine contamination. Use of the SourceTracker algorithm confirmed many of our taxonomic observations as human skin was the primary source of bacteria on restroom surfaces. Overall, these results demonstrate that restroom surfaces host relatively diverse microbial communities dominated by human-associated bacteria with clear linkages between communities on or in different body sites and those communities found on restroom surfaces. More generally, this work is relevant to the public health field as we show that human associated microbes are commonly found on restroom surfaces suggesting that bacterial pathogens could readily be transmitted between individuals by the touching of surfaces. Furthermore, we demonstrate that we can use high-throughput analyses of bacterial communities to determine sources of bacteria on indoor surfaces, an approach which could be used to track pathogen transmission and test the efficacy of hygiene practices."	ENVO:building	TCAG	Microbial biogeography of public restroom surfaces	NA	0	"Ten surfaces (door handles into and out of the restroom, handles into and out of a restroom stall, faucet handles, soap dispenser, toilet seat, toilet flush handle, floor around the toilet and floor around the sink) in six male and six female restrooms evenly distributed across two buildings on the University of Colorado at Boulder campus were sampled on a single day in November 2010. Surfaces where sampled using sterile, cotton-tipped swabs as described previously?[14],?[15]. As the 12 restrooms were nearly identical in design, we were able to swab the same area at each location between restrooms. In order to characterize tap water communities as a potential source of bacteria, 1 L of faucet water from six of the restrooms (each building having the same water source for each restroom sampled) was collected and filtered through 0.2 ?m bottle top filters (Nalgene, Rochester, NY, USA)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	22132229	NA	NA	NA	Gilberto Flores	CCME	GAZ:United States of America	NA	NA	0	NA	NA	NA	NA	n	NA	Noah Fierer	Flores_restroom_surface_biogeography	pyrosequencing	NA	NA	NA	restroom surface metagenome	3	ENVO:city	Titanium	NA	NA	NA	NA	NA	flores_restroom_surface_biogeography	restroom surface biogeography	256318	CCME	NA	NA	NA	1335	"Genomic DNA was extracted from the swabs and filters using the MO BIO PowerSoil DNA isolation kit following the manufacturer's protocol with the modifications of Fierer?et al.?[14]. A portion of the 16 S rRNA gene spanning the V1_V2 regions was amplified using the primer set (27 F/338R), PCR mixture conditions and thermal cycling conditions described in Fierer?et al.?[15]. PCR amplicons of triplicate reactions for each sample were pooled at approximately equal amounts and pyrosequenced at 454 Life Sciences (Branford, CT, USA) on their GS Junior system. A total of 337,333 high-quality partial 16 S rRNA gene sequences were obtained from 101 of the 120 surface samples collected, averaging approximately 3,340 sequences per sample (ranging from 513_6,771) (Table S1) in 4 GS Junior runs, with the best run containing 116,004 high-quality reads. An additional 16,416 sequences (ranging from 2161_5084 per sample) were generated for five of the six water samples collected for source tracking analysis. Each sample was amplified with a unique barcode to enable multiplexing in the GS Junior runs. The barcoded sequencing reads can be separated by data analysis software providing high confidence in assigning sequencing read to each sample. Sequence data generated as part of this study is available upon request by contacting the corresponding author."	NA	NA	y	NA	NA	ENVO:surface	16S rRNA	Ekeley	NA	V2	CCME	NA	Door in	NA	NA	NA	40.01499	NA	NA	Door handle area into restroom
PTAF1.489471	GCAGTTCATATC	YATGCTGCCTCCCGTAGGAGT	jesse.stombaugh@colorado.edu	NA	restroom surface	Door_handle_area_into_restroom	NA	NA	NA	n	12/13/10	restroom surface metagenome	NA	GFL0IUQ	NA	mimarks-survey	NA	NA	NA	NA	0	NA	NA	1624.1	10-Nov	Female	0	NA	NA	NA	NA	NA	NA	NA	NA	NA	NA	"1,swab"	A	NA	Microbial biogeography of public restroom surfaces	NA	NA	-105.27055	y	NA	CCME	NA	NA	NA	"We spend the majority of our lives indoors where we are constantly exposed to bacteria residing on surfaces. However, the diversity of these surface-associated communities is largely unknown. We explored the biogeographical patterns exhibited by bacteria across ten surfaces within each of twelve public restrooms. Using high-throughput barcoded pyrosequencing of the 16S rRNA gene, we identified 19 bacterial phyla across all surfaces. Most sequences belonged to four phyla: Actinobacteria, Bacteriodetes, Firmicutes and Proteobacteria. The communities clustered into three general categories: those found on surfaces associated with toilets, those on the restroom floor, and those found on surfaces routinely touched with hands. On toilet surfaces, gut-associated taxa were more prevalent, suggesting fecal contamination of these surfaces. Floor surfaces were the most diverse of all communities and contained several taxa commonly found in soils. Skin-associated bacteria, especially the Propionibacteriaceae, dominated surfaces routinely touched with our hands. Certain taxa were more common in female than in male restrooms as vagina-associated Lactobacillaceae were widely distributed in female restrooms, likely from urine contamination. Use of the SourceTracker algorithm confirmed many of our taxonomic observations as human skin was the primary source of bacteria on restroom surfaces. Overall, these results demonstrate that restroom surfaces host relatively diverse microbial communities dominated by human-associated bacteria with clear linkages between communities on or in different body sites and those communities found on restroom surfaces. More generally, this work is relevant to the public health field as we show that human associated microbes are commonly found on restroom surfaces suggesting that bacterial pathogens could readily be transmitted between individuals by the touching of surfaces. Furthermore, we demonstrate that we can use high-throughput analyses of bacterial communities to determine sources of bacteria on indoor surfaces, an approach which could be used to track pathogen transmission and test the efficacy of hygiene practices."	ENVO:building	TCAG	Microbial biogeography of public restroom surfaces	NA	0	"Ten surfaces (door handles into and out of the restroom, handles into and out of a restroom stall, faucet handles, soap dispenser, toilet seat, toilet flush handle, floor around the toilet and floor around the sink) in six male and six female restrooms evenly distributed across two buildings on the University of Colorado at Boulder campus were sampled on a single day in November 2010. Surfaces where sampled using sterile, cotton-tipped swabs as described previously?[14],?[15]. As the 12 restrooms were nearly identical in design, we were able to swab the same area at each location between restrooms. In order to characterize tap water communities as a potential source of bacteria, 1 L of faucet water from six of the restrooms (each building having the same water source for each restroom sampled) was collected and filtered through 0.2 ?m bottle top filters (Nalgene, Rochester, NY, USA)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	22132229	NA	NA	NA	Gilberto Flores	CCME	GAZ:United States of America	NA	NA	0	NA	NA	NA	NA	n	NA	Noah Fierer	Flores_restroom_surface_biogeography	pyrosequencing	NA	NA	NA	restroom surface metagenome	1	ENVO:city	Titanium	NA	NA	NA	NA	NA	flores_restroom_surface_biogeography	restroom surface biogeography	256318	CCME	NA	NA	NA	1335	"Genomic DNA was extracted from the swabs and filters using the MO BIO PowerSoil DNA isolation kit following the manufacturer's protocol with the modifications of Fierer?et al.?[14]. A portion of the 16 S rRNA gene spanning the V1_V2 regions was amplified using the primer set (27 F/338R), PCR mixture conditions and thermal cycling conditions described in Fierer?et al.?[15]. PCR amplicons of triplicate reactions for each sample were pooled at approximately equal amounts and pyrosequenced at 454 Life Sciences (Branford, CT, USA) on their GS Junior system. A total of 337,333 high-quality partial 16 S rRNA gene sequences were obtained from 101 of the 120 surface samples collected, averaging approximately 3,340 sequences per sample (ranging from 513_6,771) (Table S1) in 4 GS Junior runs, with the best run containing 116,004 high-quality reads. An additional 16,416 sequences (ranging from 2161_5084 per sample) were generated for five of the six water samples collected for source tracking analysis. Each sample was amplified with a unique barcode to enable multiplexing in the GS Junior runs. The barcoded sequencing reads can be separated by data analysis software providing high confidence in assigning sequencing read to each sample. Sequence data generated as part of this study is available upon request by contacting the corresponding author."	NA	NA	y	NA	NA	ENVO:surface	16S rRNA	Porter	NA	V2	CCME	NA	Door in	NA	NA	NA	40.01499	NA	NA	Door handle area into restroom
PTCF1.489467	ATAGGCGATCTC	YATGCTGCCTCCCGTAGGAGT	jesse.stombaugh@colorado.edu	NA	restroom surface	Door_handle_area_into_restroom	NA	NA	NA	n	12/13/10	restroom surface metagenome	NA	GFL0IUQ	NA	mimarks-survey	NA	NA	NA	NA	0	NA	NA	1624.1	10-Nov	Female	0	NA	NA	NA	NA	NA	NA	NA	NA	NA	NA	"1,swab"	C	NA	Microbial biogeography of public restroom surfaces	NA	NA	-105.27055	y	NA	CCME	NA	NA	NA	"We spend the majority of our lives indoors where we are constantly exposed to bacteria residing on surfaces. However, the diversity of these surface-associated communities is largely unknown. We explored the biogeographical patterns exhibited by bacteria across ten surfaces within each of twelve public restrooms. Using high-throughput barcoded pyrosequencing of the 16S rRNA gene, we identified 19 bacterial phyla across all surfaces. Most sequences belonged to four phyla: Actinobacteria, Bacteriodetes, Firmicutes and Proteobacteria. The communities clustered into three general categories: those found on surfaces associated with toilets, those on the restroom floor, and those found on surfaces routinely touched with hands. On toilet surfaces, gut-associated taxa were more prevalent, suggesting fecal contamination of these surfaces. Floor surfaces were the most diverse of all communities and contained several taxa commonly found in soils. Skin-associated bacteria, especially the Propionibacteriaceae, dominated surfaces routinely touched with our hands. Certain taxa were more common in female than in male restrooms as vagina-associated Lactobacillaceae were widely distributed in female restrooms, likely from urine contamination. Use of the SourceTracker algorithm confirmed many of our taxonomic observations as human skin was the primary source of bacteria on restroom surfaces. Overall, these results demonstrate that restroom surfaces host relatively diverse microbial communities dominated by human-associated bacteria with clear linkages between communities on or in different body sites and those communities found on restroom surfaces. More generally, this work is relevant to the public health field as we show that human associated microbes are commonly found on restroom surfaces suggesting that bacterial pathogens could readily be transmitted between individuals by the touching of surfaces. Furthermore, we demonstrate that we can use high-throughput analyses of bacterial communities to determine sources of bacteria on indoor surfaces, an approach which could be used to track pathogen transmission and test the efficacy of hygiene practices."	ENVO:building	TCAG	Microbial biogeography of public restroom surfaces	NA	0	"Ten surfaces (door handles into and out of the restroom, handles into and out of a restroom stall, faucet handles, soap dispenser, toilet seat, toilet flush handle, floor around the toilet and floor around the sink) in six male and six female restrooms evenly distributed across two buildings on the University of Colorado at Boulder campus were sampled on a single day in November 2010. Surfaces where sampled using sterile, cotton-tipped swabs as described previously?[14],?[15]. As the 12 restrooms were nearly identical in design, we were able to swab the same area at each location between restrooms. In order to characterize tap water communities as a potential source of bacteria, 1 L of faucet water from six of the restrooms (each building having the same water source for each restroom sampled) was collected and filtered through 0.2 ?m bottle top filters (Nalgene, Rochester, NY, USA)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	22132229	NA	NA	NA	Gilberto Flores	CCME	GAZ:United States of America	NA	NA	0	NA	NA	NA	NA	n	NA	Noah Fierer	Flores_restroom_surface_biogeography	pyrosequencing	NA	NA	NA	restroom surface metagenome	3	ENVO:city	Titanium	NA	NA	NA	NA	NA	flores_restroom_surface_biogeography	restroom surface biogeography	256318	CCME	NA	NA	NA	1335	"Genomic DNA was extracted from the swabs and filters using the MO BIO PowerSoil DNA isolation kit following the manufacturer's protocol with the modifications of Fierer?et al.?[14]. A portion of the 16 S rRNA gene spanning the V1_V2 regions was amplified using the primer set (27 F/338R), PCR mixture conditions and thermal cycling conditions described in Fierer?et al.?[15]. PCR amplicons of triplicate reactions for each sample were pooled at approximately equal amounts and pyrosequenced at 454 Life Sciences (Branford, CT, USA) on their GS Junior system. A total of 337,333 high-quality partial 16 S rRNA gene sequences were obtained from 101 of the 120 surface samples collected, averaging approximately 3,340 sequences per sample (ranging from 513_6,771) (Table S1) in 4 GS Junior runs, with the best run containing 116,004 high-quality reads. An additional 16,416 sequences (ranging from 2161_5084 per sample) were generated for five of the six water samples collected for source tracking analysis. Each sample was amplified with a unique barcode to enable multiplexing in the GS Junior runs. The barcoded sequencing reads can be separated by data analysis software providing high confidence in assigning sequencing read to each sample. Sequence data generated as part of this study is available upon request by contacting the corresponding author."	NA	NA	y	NA	NA	ENVO:surface	16S rRNA	Porter	NA	V2	CCME	NA	Door in	NA	NA	NA	40.01499	NA	NA	Door handle area into restroom
PTCM1.489536	AGTTAGTGCGTC	YATGCTGCCTCCCGTAGGAGT	jesse.stombaugh@colorado.edu	NA	restroom surface	Door_handle_area_into_restroom	NA	NA	NA	n	12/13/10	restroom surface metagenome	NA	GFL0IUQ	NA	mimarks-survey	NA	NA	NA	NA	0	NA	NA	1624.1	10-Nov	Male	0	NA	NA	NA	NA	NA	NA	NA	NA	NA	NA	"1,swab"	C	NA	Microbial biogeography of public restroom surfaces	NA	NA	-105.27055	y	NA	CCME	NA	NA	NA	"We spend the majority of our lives indoors where we are constantly exposed to bacteria residing on surfaces. However, the diversity of these surface-associated communities is largely unknown. We explored the biogeographical patterns exhibited by bacteria across ten surfaces within each of twelve public restrooms. Using high-throughput barcoded pyrosequencing of the 16S rRNA gene, we identified 19 bacterial phyla across all surfaces. Most sequences belonged to four phyla: Actinobacteria, Bacteriodetes, Firmicutes and Proteobacteria. The communities clustered into three general categories: those found on surfaces associated with toilets, those on the restroom floor, and those found on surfaces routinely touched with hands. On toilet surfaces, gut-associated taxa were more prevalent, suggesting fecal contamination of these surfaces. Floor surfaces were the most diverse of all communities and contained several taxa commonly found in soils. Skin-associated bacteria, especially the Propionibacteriaceae, dominated surfaces routinely touched with our hands. Certain taxa were more common in female than in male restrooms as vagina-associated Lactobacillaceae were widely distributed in female restrooms, likely from urine contamination. Use of the SourceTracker algorithm confirmed many of our taxonomic observations as human skin was the primary source of bacteria on restroom surfaces. Overall, these results demonstrate that restroom surfaces host relatively diverse microbial communities dominated by human-associated bacteria with clear linkages between communities on or in different body sites and those communities found on restroom surfaces. More generally, this work is relevant to the public health field as we show that human associated microbes are commonly found on restroom surfaces suggesting that bacterial pathogens could readily be transmitted between individuals by the touching of surfaces. Furthermore, we demonstrate that we can use high-throughput analyses of bacterial communities to determine sources of bacteria on indoor surfaces, an approach which could be used to track pathogen transmission and test the efficacy of hygiene practices."	ENVO:building	TCAG	Microbial biogeography of public restroom surfaces	NA	0	"Ten surfaces (door handles into and out of the restroom, handles into and out of a restroom stall, faucet handles, soap dispenser, toilet seat, toilet flush handle, floor around the toilet and floor around the sink) in six male and six female restrooms evenly distributed across two buildings on the University of Colorado at Boulder campus were sampled on a single day in November 2010. Surfaces where sampled using sterile, cotton-tipped swabs as described previously?[14],?[15]. As the 12 restrooms were nearly identical in design, we were able to swab the same area at each location between restrooms. In order to characterize tap water communities as a potential source of bacteria, 1 L of faucet water from six of the restrooms (each building having the same water source for each restroom sampled) was collected and filtered through 0.2 ?m bottle top filters (Nalgene, Rochester, NY, USA)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	22132229	NA	NA	NA	Gilberto Flores	CCME	GAZ:United States of America	NA	NA	0	NA	NA	NA	NA	n	NA	Noah Fierer	Flores_restroom_surface_biogeography	pyrosequencing	NA	NA	NA	restroom surface metagenome	3	ENVO:city	Titanium	NA	NA	NA	NA	NA	flores_restroom_surface_biogeography	restroom surface biogeography	256318	CCME	NA	NA	NA	1335	"Genomic DNA was extracted from the swabs and filters using the MO BIO PowerSoil DNA isolation kit following the manufacturer's protocol with the modifications of Fierer?et al.?[14]. A portion of the 16 S rRNA gene spanning the V1_V2 regions was amplified using the primer set (27 F/338R), PCR mixture conditions and thermal cycling conditions described in Fierer?et al.?[15]. PCR amplicons of triplicate reactions for each sample were pooled at approximately equal amounts and pyrosequenced at 454 Life Sciences (Branford, CT, USA) on their GS Junior system. A total of 337,333 high-quality partial 16 S rRNA gene sequences were obtained from 101 of the 120 surface samples collected, averaging approximately 3,340 sequences per sample (ranging from 513_6,771) (Table S1) in 4 GS Junior runs, with the best run containing 116,004 high-quality reads. An additional 16,416 sequences (ranging from 2161_5084 per sample) were generated for five of the six water samples collected for source tracking analysis. Each sample was amplified with a unique barcode to enable multiplexing in the GS Junior runs. The barcoded sequencing reads can be separated by data analysis software providing high confidence in assigning sequencing read to each sample. Sequence data generated as part of this study is available upon request by contacting the corresponding author."	NA	NA	y	NA	NA	ENVO:surface	16S rRNA	Porter	NA	V2	CCME	NA	Door in	NA	NA	NA	40.01499	NA	NA	Door handle area into restroom
EKAF2.489569	GACAGCGTTGAC	YATGCTGCCTCCCGTAGGAGT	jesse.stombaugh@colorado.edu	NA	restroom surface	Door_handle_area_out_of_restroom	NA	NA	NA	n	12/13/10	restroom surface metagenome	NA	GFL0IUQ	NA	mimarks-survey	NA	NA	NA	NA	0	NA	NA	1624.1	10-Nov	Female	0	NA	NA	NA	NA	NA	NA	NA	NA	NA	NA	"1,swab"	A	NA	Microbial biogeography of public restroom surfaces	NA	NA	-105.27055	y	NA	CCME	NA	NA	NA	"We spend the majority of our lives indoors where we are constantly exposed to bacteria residing on surfaces. However, the diversity of these surface-associated communities is largely unknown. We explored the biogeographical patterns exhibited by bacteria across ten surfaces within each of twelve public restrooms. Using high-throughput barcoded pyrosequencing of the 16S rRNA gene, we identified 19 bacterial phyla across all surfaces. Most sequences belonged to four phyla: Actinobacteria, Bacteriodetes, Firmicutes and Proteobacteria. The communities clustered into three general categories: those found on surfaces associated with toilets, those on the restroom floor, and those found on surfaces routinely touched with hands. On toilet surfaces, gut-associated taxa were more prevalent, suggesting fecal contamination of these surfaces. Floor surfaces were the most diverse of all communities and contained several taxa commonly found in soils. Skin-associated bacteria, especially the Propionibacteriaceae, dominated surfaces routinely touched with our hands. Certain taxa were more common in female than in male restrooms as vagina-associated Lactobacillaceae were widely distributed in female restrooms, likely from urine contamination. Use of the SourceTracker algorithm confirmed many of our taxonomic observations as human skin was the primary source of bacteria on restroom surfaces. Overall, these results demonstrate that restroom surfaces host relatively diverse microbial communities dominated by human-associated bacteria with clear linkages between communities on or in different body sites and those communities found on restroom surfaces. More generally, this work is relevant to the public health field as we show that human associated microbes are commonly found on restroom surfaces suggesting that bacterial pathogens could readily be transmitted between individuals by the touching of surfaces. Furthermore, we demonstrate that we can use high-throughput analyses of bacterial communities to determine sources of bacteria on indoor surfaces, an approach which could be used to track pathogen transmission and test the efficacy of hygiene practices."	ENVO:building	TCAG	Microbial biogeography of public restroom surfaces	NA	0	"Ten surfaces (door handles into and out of the restroom, handles into and out of a restroom stall, faucet handles, soap dispenser, toilet seat, toilet flush handle, floor around the toilet and floor around the sink) in six male and six female restrooms evenly distributed across two buildings on the University of Colorado at Boulder campus were sampled on a single day in November 2010. Surfaces where sampled using sterile, cotton-tipped swabs as described previously?[14],?[15]. As the 12 restrooms were nearly identical in design, we were able to swab the same area at each location between restrooms. In order to characterize tap water communities as a potential source of bacteria, 1 L of faucet water from six of the restrooms (each building having the same water source for each restroom sampled) was collected and filtered through 0.2 ?m bottle top filters (Nalgene, Rochester, NY, USA)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	22132229	NA	NA	NA	Gilberto Flores	CCME	GAZ:United States of America	NA	NA	0	NA	NA	NA	NA	n	NA	Noah Fierer	Flores_restroom_surface_biogeography	pyrosequencing	NA	NA	NA	restroom surface metagenome	1	ENVO:city	Titanium	NA	NA	NA	NA	NA	flores_restroom_surface_biogeography	restroom surface biogeography	256318	CCME	NA	NA	NA	1335	"Genomic DNA was extracted from the swabs and filters using the MO BIO PowerSoil DNA isolation kit following the manufacturer's protocol with the modifications of Fierer?et al.?[14]. A portion of the 16 S rRNA gene spanning the V1_V2 regions was amplified using the primer set (27 F/338R), PCR mixture conditions and thermal cycling conditions described in Fierer?et al.?[15]. PCR amplicons of triplicate reactions for each sample were pooled at approximately equal amounts and pyrosequenced at 454 Life Sciences (Branford, CT, USA) on their GS Junior system. A total of 337,333 high-quality partial 16 S rRNA gene sequences were obtained from 101 of the 120 surface samples collected, averaging approximately 3,340 sequences per sample (ranging from 513_6,771) (Table S1) in 4 GS Junior runs, with the best run containing 116,004 high-quality reads. An additional 16,416 sequences (ranging from 2161_5084 per sample) were generated for five of the six water samples collected for source tracking analysis. Each sample was amplified with a unique barcode to enable multiplexing in the GS Junior runs. The barcoded sequencing reads can be separated by data analysis software providing high confidence in assigning sequencing read to each sample. Sequence data generated as part of this study is available upon request by contacting the corresponding author."	NA	NA	y	NA	NA	ENVO:surface	16S rRNA	Ekeley	NA	V2	CCME	NA	Door out	NA	NA	NA	40.01499	NA	NA	Door handle area out of restroom
EKBF2.489510	GAGACAGCTTGC	YATGCTGCCTCCCGTAGGAGT	jesse.stombaugh@colorado.edu	NA	restroom surface	Door_handle_area_out_of_restroom	NA	NA	NA	n	12/13/10	restroom surface metagenome	NA	GFL0IUQ	NA	mimarks-survey	NA	NA	NA	NA	0	NA	NA	1624.1	10-Nov	Female	0	NA	NA	NA	NA	NA	NA	NA	NA	NA	NA	"1,swab"	B	NA	Microbial biogeography of public restroom surfaces	NA	NA	-105.27055	y	NA	CCME	NA	NA	NA	"We spend the majority of our lives indoors where we are constantly exposed to bacteria residing on surfaces. However, the diversity of these surface-associated communities is largely unknown. We explored the biogeographical patterns exhibited by bacteria across ten surfaces within each of twelve public restrooms. Using high-throughput barcoded pyrosequencing of the 16S rRNA gene, we identified 19 bacterial phyla across all surfaces. Most sequences belonged to four phyla: Actinobacteria, Bacteriodetes, Firmicutes and Proteobacteria. The communities clustered into three general categories: those found on surfaces associated with toilets, those on the restroom floor, and those found on surfaces routinely touched with hands. On toilet surfaces, gut-associated taxa were more prevalent, suggesting fecal contamination of these surfaces. Floor surfaces were the most diverse of all communities and contained several taxa commonly found in soils. Skin-associated bacteria, especially the Propionibacteriaceae, dominated surfaces routinely touched with our hands. Certain taxa were more common in female than in male restrooms as vagina-associated Lactobacillaceae were widely distributed in female restrooms, likely from urine contamination. Use of the SourceTracker algorithm confirmed many of our taxonomic observations as human skin was the primary source of bacteria on restroom surfaces. Overall, these results demonstrate that restroom surfaces host relatively diverse microbial communities dominated by human-associated bacteria with clear linkages between communities on or in different body sites and those communities found on restroom surfaces. More generally, this work is relevant to the public health field as we show that human associated microbes are commonly found on restroom surfaces suggesting that bacterial pathogens could readily be transmitted between individuals by the touching of surfaces. Furthermore, we demonstrate that we can use high-throughput analyses of bacterial communities to determine sources of bacteria on indoor surfaces, an approach which could be used to track pathogen transmission and test the efficacy of hygiene practices."	ENVO:building	TCAG	Microbial biogeography of public restroom surfaces	NA	0	"Ten surfaces (door handles into and out of the restroom, handles into and out of a restroom stall, faucet handles, soap dispenser, toilet seat, toilet flush handle, floor around the toilet and floor around the sink) in six male and six female restrooms evenly distributed across two buildings on the University of Colorado at Boulder campus were sampled on a single day in November 2010. Surfaces where sampled using sterile, cotton-tipped swabs as described previously?[14],?[15]. As the 12 restrooms were nearly identical in design, we were able to swab the same area at each location between restrooms. In order to characterize tap water communities as a potential source of bacteria, 1 L of faucet water from six of the restrooms (each building having the same water source for each restroom sampled) was collected and filtered through 0.2 ?m bottle top filters (Nalgene, Rochester, NY, USA)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	22132229	NA	NA	NA	Gilberto Flores	CCME	GAZ:United States of America	NA	NA	0	NA	NA	NA	NA	n	NA	Noah Fierer	Flores_restroom_surface_biogeography	pyrosequencing	NA	NA	NA	restroom surface metagenome	2	ENVO:city	Titanium	NA	NA	NA	NA	NA	flores_restroom_surface_biogeography	restroom surface biogeography	256318	CCME	NA	NA	NA	1335	"Genomic DNA was extracted from the swabs and filters using the MO BIO PowerSoil DNA isolation kit following the manufacturer's protocol with the modifications of Fierer?et al.?[14]. A portion of the 16 S rRNA gene spanning the V1_V2 regions was amplified using the primer set (27 F/338R), PCR mixture conditions and thermal cycling conditions described in Fierer?et al.?[15]. PCR amplicons of triplicate reactions for each sample were pooled at approximately equal amounts and pyrosequenced at 454 Life Sciences (Branford, CT, USA) on their GS Junior system. A total of 337,333 high-quality partial 16 S rRNA gene sequences were obtained from 101 of the 120 surface samples collected, averaging approximately 3,340 sequences per sample (ranging from 513_6,771) (Table S1) in 4 GS Junior runs, with the best run containing 116,004 high-quality reads. An additional 16,416 sequences (ranging from 2161_5084 per sample) were generated for five of the six water samples collected for source tracking analysis. Each sample was amplified with a unique barcode to enable multiplexing in the GS Junior runs. The barcoded sequencing reads can be separated by data analysis software providing high confidence in assigning sequencing read to each sample. Sequence data generated as part of this study is available upon request by contacting the corresponding author."	NA	NA	y	NA	NA	ENVO:surface	16S rRNA	Ekeley	NA	V2	CCME	NA	Door out	NA	NA	NA	40.01499	NA	NA	Door handle area out of restroom
EKBM2.489466	GACGTTGCACAG	YATGCTGCCTCCCGTAGGAGT	jesse.stombaugh@colorado.edu	NA	restroom surface	Door_handle_area_out_of_restroom	NA	NA	NA	n	12/13/10	restroom surface metagenome	NA	GFL0IUQ	NA	mimarks-survey	NA	NA	NA	NA	0	NA	NA	1624.1	10-Nov	Male	0	NA	NA	NA	NA	NA	NA	NA	NA	NA	NA	"1,swab"	B	NA	Microbial biogeography of public restroom surfaces	NA	NA	-105.27055	y	NA	CCME	NA	NA	NA	"We spend the majority of our lives indoors where we are constantly exposed to bacteria residing on surfaces. However, the diversity of these surface-associated communities is largely unknown. We explored the biogeographical patterns exhibited by bacteria across ten surfaces within each of twelve public restrooms. Using high-throughput barcoded pyrosequencing of the 16S rRNA gene, we identified 19 bacterial phyla across all surfaces. Most sequences belonged to four phyla: Actinobacteria, Bacteriodetes, Firmicutes and Proteobacteria. The communities clustered into three general categories: those found on surfaces associated with toilets, those on the restroom floor, and those found on surfaces routinely touched with hands. On toilet surfaces, gut-associated taxa were more prevalent, suggesting fecal contamination of these surfaces. Floor surfaces were the most diverse of all communities and contained several taxa commonly found in soils. Skin-associated bacteria, especially the Propionibacteriaceae, dominated surfaces routinely touched with our hands. Certain taxa were more common in female than in male restrooms as vagina-associated Lactobacillaceae were widely distributed in female restrooms, likely from urine contamination. Use of the SourceTracker algorithm confirmed many of our taxonomic observations as human skin was the primary source of bacteria on restroom surfaces. Overall, these results demonstrate that restroom surfaces host relatively diverse microbial communities dominated by human-associated bacteria with clear linkages between communities on or in different body sites and those communities found on restroom surfaces. More generally, this work is relevant to the public health field as we show that human associated microbes are commonly found on restroom surfaces suggesting that bacterial pathogens could readily be transmitted between individuals by the touching of surfaces. Furthermore, we demonstrate that we can use high-throughput analyses of bacterial communities to determine sources of bacteria on indoor surfaces, an approach which could be used to track pathogen transmission and test the efficacy of hygiene practices."	ENVO:building	TCAG	Microbial biogeography of public restroom surfaces	NA	0	"Ten surfaces (door handles into and out of the restroom, handles into and out of a restroom stall, faucet handles, soap dispenser, toilet seat, toilet flush handle, floor around the toilet and floor around the sink) in six male and six female restrooms evenly distributed across two buildings on the University of Colorado at Boulder campus were sampled on a single day in November 2010. Surfaces where sampled using sterile, cotton-tipped swabs as described previously?[14],?[15]. As the 12 restrooms were nearly identical in design, we were able to swab the same area at each location between restrooms. In order to characterize tap water communities as a potential source of bacteria, 1 L of faucet water from six of the restrooms (each building having the same water source for each restroom sampled) was collected and filtered through 0.2 ?m bottle top filters (Nalgene, Rochester, NY, USA)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	22132229	NA	NA	NA	Gilberto Flores	CCME	GAZ:United States of America	NA	NA	0	NA	NA	NA	NA	n	NA	Noah Fierer	Flores_restroom_surface_biogeography	pyrosequencing	NA	NA	NA	restroom surface metagenome	2	ENVO:city	Titanium	NA	NA	NA	NA	NA	flores_restroom_surface_biogeography	restroom surface biogeography	256318	CCME	NA	NA	NA	1335	"Genomic DNA was extracted from the swabs and filters using the MO BIO PowerSoil DNA isolation kit following the manufacturer's protocol with the modifications of Fierer?et al.?[14]. A portion of the 16 S rRNA gene spanning the V1_V2 regions was amplified using the primer set (27 F/338R), PCR mixture conditions and thermal cycling conditions described in Fierer?et al.?[15]. PCR amplicons of triplicate reactions for each sample were pooled at approximately equal amounts and pyrosequenced at 454 Life Sciences (Branford, CT, USA) on their GS Junior system. A total of 337,333 high-quality partial 16 S rRNA gene sequences were obtained from 101 of the 120 surface samples collected, averaging approximately 3,340 sequences per sample (ranging from 513_6,771) (Table S1) in 4 GS Junior runs, with the best run containing 116,004 high-quality reads. An additional 16,416 sequences (ranging from 2161_5084 per sample) were generated for five of the six water samples collected for source tracking analysis. Each sample was amplified with a unique barcode to enable multiplexing in the GS Junior runs. The barcoded sequencing reads can be separated by data analysis software providing high confidence in assigning sequencing read to each sample. Sequence data generated as part of this study is available upon request by contacting the corresponding author."	NA	NA	y	NA	NA	ENVO:surface	16S rRNA	Ekeley	NA	V2	CCME	NA	Door out	NA	NA	NA	40.01499	NA	NA	Door handle area out of restroom
EKCF2.489571	GATCGTCCAGAT	YATGCTGCCTCCCGTAGGAGT	jesse.stombaugh@colorado.edu	NA	restroom surface	Door_handle_area_out_of_restroom	NA	NA	NA	n	12/13/10	restroom surface metagenome	NA	GFL0IUQ	NA	mimarks-survey	NA	NA	NA	NA	0	NA	NA	1624.1	10-Nov	Female	0	NA	NA	NA	NA	NA	NA	NA	NA	NA	NA	"1,swab"	C	NA	Microbial biogeography of public restroom surfaces	NA	NA	-105.27055	y	NA	CCME	NA	NA	NA	"We spend the majority of our lives indoors where we are constantly exposed to bacteria residing on surfaces. However, the diversity of these surface-associated communities is largely unknown. We explored the biogeographical patterns exhibited by bacteria across ten surfaces within each of twelve public restrooms. Using high-throughput barcoded pyrosequencing of the 16S rRNA gene, we identified 19 bacterial phyla across all surfaces. Most sequences belonged to four phyla: Actinobacteria, Bacteriodetes, Firmicutes and Proteobacteria. The communities clustered into three general categories: those found on surfaces associated with toilets, those on the restroom floor, and those found on surfaces routinely touched with hands. On toilet surfaces, gut-associated taxa were more prevalent, suggesting fecal contamination of these surfaces. Floor surfaces were the most diverse of all communities and contained several taxa commonly found in soils. Skin-associated bacteria, especially the Propionibacteriaceae, dominated surfaces routinely touched with our hands. Certain taxa were more common in female than in male restrooms as vagina-associated Lactobacillaceae were widely distributed in female restrooms, likely from urine contamination. Use of the SourceTracker algorithm confirmed many of our taxonomic observations as human skin was the primary source of bacteria on restroom surfaces. Overall, these results demonstrate that restroom surfaces host relatively diverse microbial communities dominated by human-associated bacteria with clear linkages between communities on or in different body sites and those communities found on restroom surfaces. More generally, this work is relevant to the public health field as we show that human associated microbes are commonly found on restroom surfaces suggesting that bacterial pathogens could readily be transmitted between individuals by the touching of surfaces. Furthermore, we demonstrate that we can use high-throughput analyses of bacterial communities to determine sources of bacteria on indoor surfaces, an approach which could be used to track pathogen transmission and test the efficacy of hygiene practices."	ENVO:building	TCAG	Microbial biogeography of public restroom surfaces	NA	0	"Ten surfaces (door handles into and out of the restroom, handles into and out of a restroom stall, faucet handles, soap dispenser, toilet seat, toilet flush handle, floor around the toilet and floor around the sink) in six male and six female restrooms evenly distributed across two buildings on the University of Colorado at Boulder campus were sampled on a single day in November 2010. Surfaces where sampled using sterile, cotton-tipped swabs as described previously?[14],?[15]. As the 12 restrooms were nearly identical in design, we were able to swab the same area at each location between restrooms. In order to characterize tap water communities as a potential source of bacteria, 1 L of faucet water from six of the restrooms (each building having the same water source for each restroom sampled) was collected and filtered through 0.2 ?m bottle top filters (Nalgene, Rochester, NY, USA)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	22132229	NA	NA	NA	Gilberto Flores	CCME	GAZ:United States of America	NA	NA	0	NA	NA	NA	NA	n	NA	Noah Fierer	Flores_restroom_surface_biogeography	pyrosequencing	NA	NA	NA	restroom surface metagenome	3	ENVO:city	Titanium	NA	NA	NA	NA	NA	flores_restroom_surface_biogeography	restroom surface biogeography	256318	CCME	NA	NA	NA	1335	"Genomic DNA was extracted from the swabs and filters using the MO BIO PowerSoil DNA isolation kit following the manufacturer's protocol with the modifications of Fierer?et al.?[14]. A portion of the 16 S rRNA gene spanning the V1_V2 regions was amplified using the primer set (27 F/338R), PCR mixture conditions and thermal cycling conditions described in Fierer?et al.?[15]. PCR amplicons of triplicate reactions for each sample were pooled at approximately equal amounts and pyrosequenced at 454 Life Sciences (Branford, CT, USA) on their GS Junior system. A total of 337,333 high-quality partial 16 S rRNA gene sequences were obtained from 101 of the 120 surface samples collected, averaging approximately 3,340 sequences per sample (ranging from 513_6,771) (Table S1) in 4 GS Junior runs, with the best run containing 116,004 high-quality reads. An additional 16,416 sequences (ranging from 2161_5084 per sample) were generated for five of the six water samples collected for source tracking analysis. Each sample was amplified with a unique barcode to enable multiplexing in the GS Junior runs. The barcoded sequencing reads can be separated by data analysis software providing high confidence in assigning sequencing read to each sample. Sequence data generated as part of this study is available upon request by contacting the corresponding author."	NA	NA	y	NA	NA	ENVO:surface	16S rRNA	Ekeley	NA	V2	CCME	NA	Door out	NA	NA	NA	40.01499	NA	NA	Door handle area out of restroom
EKCM2.489495	GAGTCTGAGTCT	YATGCTGCCTCCCGTAGGAGT	jesse.stombaugh@colorado.edu	NA	restroom surface	Door_handle_area_out_of_restroom	NA	NA	NA	n	12/13/10	restroom surface metagenome	NA	GFL0IUQ	NA	mimarks-survey	NA	NA	NA	NA	0	NA	NA	1624.1	10-Nov	Male	0	NA	NA	NA	NA	NA	NA	NA	NA	NA	NA	"1,swab"	C	NA	Microbial biogeography of public restroom surfaces	NA	NA	-105.27055	y	NA	CCME	NA	NA	NA	"We spend the majority of our lives indoors where we are constantly exposed to bacteria residing on surfaces. However, the diversity of these surface-associated communities is largely unknown. We explored the biogeographical patterns exhibited by bacteria across ten surfaces within each of twelve public restrooms. Using high-throughput barcoded pyrosequencing of the 16S rRNA gene, we identified 19 bacterial phyla across all surfaces. Most sequences belonged to four phyla: Actinobacteria, Bacteriodetes, Firmicutes and Proteobacteria. The communities clustered into three general categories: those found on surfaces associated with toilets, those on the restroom floor, and those found on surfaces routinely touched with hands. On toilet surfaces, gut-associated taxa were more prevalent, suggesting fecal contamination of these surfaces. Floor surfaces were the most diverse of all communities and contained several taxa commonly found in soils. Skin-associated bacteria, especially the Propionibacteriaceae, dominated surfaces routinely touched with our hands. Certain taxa were more common in female than in male restrooms as vagina-associated Lactobacillaceae were widely distributed in female restrooms, likely from urine contamination. Use of the SourceTracker algorithm confirmed many of our taxonomic observations as human skin was the primary source of bacteria on restroom surfaces. Overall, these results demonstrate that restroom surfaces host relatively diverse microbial communities dominated by human-associated bacteria with clear linkages between communities on or in different body sites and those communities found on restroom surfaces. More generally, this work is relevant to the public health field as we show that human associated microbes are commonly found on restroom surfaces suggesting that bacterial pathogens could readily be transmitted between individuals by the touching of surfaces. Furthermore, we demonstrate that we can use high-throughput analyses of bacterial communities to determine sources of bacteria on indoor surfaces, an approach which could be used to track pathogen transmission and test the efficacy of hygiene practices."	ENVO:building	TCAG	Microbial biogeography of public restroom surfaces	NA	0	"Ten surfaces (door handles into and out of the restroom, handles into and out of a restroom stall, faucet handles, soap dispenser, toilet seat, toilet flush handle, floor around the toilet and floor around the sink) in six male and six female restrooms evenly distributed across two buildings on the University of Colorado at Boulder campus were sampled on a single day in November 2010. Surfaces where sampled using sterile, cotton-tipped swabs as described previously?[14],?[15]. As the 12 restrooms were nearly identical in design, we were able to swab the same area at each location between restrooms. In order to characterize tap water communities as a potential source of bacteria, 1 L of faucet water from six of the restrooms (each building having the same water source for each restroom sampled) was collected and filtered through 0.2 ?m bottle top filters (Nalgene, Rochester, NY, USA)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	22132229	NA	NA	NA	Gilberto Flores	CCME	GAZ:United States of America	NA	NA	0	NA	NA	NA	NA	n	NA	Noah Fierer	Flores_restroom_surface_biogeography	pyrosequencing	NA	NA	NA	restroom surface metagenome	3	ENVO:city	Titanium	NA	NA	NA	NA	NA	flores_restroom_surface_biogeography	restroom surface biogeography	256318	CCME	NA	NA	NA	1335	"Genomic DNA was extracted from the swabs and filters using the MO BIO PowerSoil DNA isolation kit following the manufacturer's protocol with the modifications of Fierer?et al.?[14]. A portion of the 16 S rRNA gene spanning the V1_V2 regions was amplified using the primer set (27 F/338R), PCR mixture conditions and thermal cycling conditions described in Fierer?et al.?[15]. PCR amplicons of triplicate reactions for each sample were pooled at approximately equal amounts and pyrosequenced at 454 Life Sciences (Branford, CT, USA) on their GS Junior system. A total of 337,333 high-quality partial 16 S rRNA gene sequences were obtained from 101 of the 120 surface samples collected, averaging approximately 3,340 sequences per sample (ranging from 513_6,771) (Table S1) in 4 GS Junior runs, with the best run containing 116,004 high-quality reads. An additional 16,416 sequences (ranging from 2161_5084 per sample) were generated for five of the six water samples collected for source tracking analysis. Each sample was amplified with a unique barcode to enable multiplexing in the GS Junior runs. The barcoded sequencing reads can be separated by data analysis software providing high confidence in assigning sequencing read to each sample. Sequence data generated as part of this study is available upon request by contacting the corresponding author."	NA	NA	y	NA	NA	ENVO:surface	16S rRNA	Ekeley	NA	V2	CCME	NA	Door out	NA	NA	NA	40.01499	NA	NA	Door handle area out of restroom
PTAF2.489522	GCATAGTAGCCG	YATGCTGCCTCCCGTAGGAGT	jesse.stombaugh@colorado.edu	NA	restroom surface	Door_handle_area_out_of_restroom	NA	NA	NA	n	12/13/10	restroom surface metagenome	NA	GFL0IUQ	NA	mimarks-survey	NA	NA	NA	NA	0	NA	NA	1624.1	10-Nov	Female	0	NA	NA	NA	NA	NA	NA	NA	NA	NA	NA	"1,swab"	A	NA	Microbial biogeography of public restroom surfaces	NA	NA	-105.27055	y	NA	CCME	NA	NA	NA	"We spend the majority of our lives indoors where we are constantly exposed to bacteria residing on surfaces. However, the diversity of these surface-associated communities is largely unknown. We explored the biogeographical patterns exhibited by bacteria across ten surfaces within each of twelve public restrooms. Using high-throughput barcoded pyrosequencing of the 16S rRNA gene, we identified 19 bacterial phyla across all surfaces. Most sequences belonged to four phyla: Actinobacteria, Bacteriodetes, Firmicutes and Proteobacteria. The communities clustered into three general categories: those found on surfaces associated with toilets, those on the restroom floor, and those found on surfaces routinely touched with hands. On toilet surfaces, gut-associated taxa were more prevalent, suggesting fecal contamination of these surfaces. Floor surfaces were the most diverse of all communities and contained several taxa commonly found in soils. Skin-associated bacteria, especially the Propionibacteriaceae, dominated surfaces routinely touched with our hands. Certain taxa were more common in female than in male restrooms as vagina-associated Lactobacillaceae were widely distributed in female restrooms, likely from urine contamination. Use of the SourceTracker algorithm confirmed many of our taxonomic observations as human skin was the primary source of bacteria on restroom surfaces. Overall, these results demonstrate that restroom surfaces host relatively diverse microbial communities dominated by human-associated bacteria with clear linkages between communities on or in different body sites and those communities found on restroom surfaces. More generally, this work is relevant to the public health field as we show that human associated microbes are commonly found on restroom surfaces suggesting that bacterial pathogens could readily be transmitted between individuals by the touching of surfaces. Furthermore, we demonstrate that we can use high-throughput analyses of bacterial communities to determine sources of bacteria on indoor surfaces, an approach which could be used to track pathogen transmission and test the efficacy of hygiene practices."	ENVO:building	TCAG	Microbial biogeography of public restroom surfaces	NA	0	"Ten surfaces (door handles into and out of the restroom, handles into and out of a restroom stall, faucet handles, soap dispenser, toilet seat, toilet flush handle, floor around the toilet and floor around the sink) in six male and six female restrooms evenly distributed across two buildings on the University of Colorado at Boulder campus were sampled on a single day in November 2010. Surfaces where sampled using sterile, cotton-tipped swabs as described previously?[14],?[15]. As the 12 restrooms were nearly identical in design, we were able to swab the same area at each location between restrooms. In order to characterize tap water communities as a potential source of bacteria, 1 L of faucet water from six of the restrooms (each building having the same water source for each restroom sampled) was collected and filtered through 0.2 ?m bottle top filters (Nalgene, Rochester, NY, USA)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	22132229	NA	NA	NA	Gilberto Flores	CCME	GAZ:United States of America	NA	NA	0	NA	NA	NA	NA	n	NA	Noah Fierer	Flores_restroom_surface_biogeography	pyrosequencing	NA	NA	NA	restroom surface metagenome	1	ENVO:city	Titanium	NA	NA	NA	NA	NA	flores_restroom_surface_biogeography	restroom surface biogeography	256318	CCME	NA	NA	NA	1335	"Genomic DNA was extracted from the swabs and filters using the MO BIO PowerSoil DNA isolation kit following the manufacturer's protocol with the modifications of Fierer?et al.?[14]. A portion of the 16 S rRNA gene spanning the V1_V2 regions was amplified using the primer set (27 F/338R), PCR mixture conditions and thermal cycling conditions described in Fierer?et al.?[15]. PCR amplicons of triplicate reactions for each sample were pooled at approximately equal amounts and pyrosequenced at 454 Life Sciences (Branford, CT, USA) on their GS Junior system. A total of 337,333 high-quality partial 16 S rRNA gene sequences were obtained from 101 of the 120 surface samples collected, averaging approximately 3,340 sequences per sample (ranging from 513_6,771) (Table S1) in 4 GS Junior runs, with the best run containing 116,004 high-quality reads. An additional 16,416 sequences (ranging from 2161_5084 per sample) were generated for five of the six water samples collected for source tracking analysis. Each sample was amplified with a unique barcode to enable multiplexing in the GS Junior runs. The barcoded sequencing reads can be separated by data analysis software providing high confidence in assigning sequencing read to each sample. Sequence data generated as part of this study is available upon request by contacting the corresponding author."	NA	NA	y	NA	NA	ENVO:surface	16S rRNA	Porter	NA	V2	CCME	NA	Door out	NA	NA	NA	40.01499	NA	NA	Door handle area out of restroom
PTAM2.489554	GCACATCGAGCA	YATGCTGCCTCCCGTAGGAGT	jesse.stombaugh@colorado.edu	NA	restroom surface	Door_handle_area_out_of_restroom	NA	NA	NA	n	12/13/10	restroom surface metagenome	NA	GFL0IUQ	NA	mimarks-survey	NA	NA	NA	NA	0	NA	NA	1624.1	10-Nov	Male	0	NA	NA	NA	NA	NA	NA	NA	NA	NA	NA	"1,swab"	A	NA	Microbial biogeography of public restroom surfaces	NA	NA	-105.27055	y	NA	CCME	NA	NA	NA	"We spend the majority of our lives indoors where we are constantly exposed to bacteria residing on surfaces. However, the diversity of these surface-associated communities is largely unknown. We explored the biogeographical patterns exhibited by bacteria across ten surfaces within each of twelve public restrooms. Using high-throughput barcoded pyrosequencing of the 16S rRNA gene, we identified 19 bacterial phyla across all surfaces. Most sequences belonged to four phyla: Actinobacteria, Bacteriodetes, Firmicutes and Proteobacteria. The communities clustered into three general categories: those found on surfaces associated with toilets, those on the restroom floor, and those found on surfaces routinely touched with hands. On toilet surfaces, gut-associated taxa were more prevalent, suggesting fecal contamination of these surfaces. Floor surfaces were the most diverse of all communities and contained several taxa commonly found in soils. Skin-associated bacteria, especially the Propionibacteriaceae, dominated surfaces routinely touched with our hands. Certain taxa were more common in female than in male restrooms as vagina-associated Lactobacillaceae were widely distributed in female restrooms, likely from urine contamination. Use of the SourceTracker algorithm confirmed many of our taxonomic observations as human skin was the primary source of bacteria on restroom surfaces. Overall, these results demonstrate that restroom surfaces host relatively diverse microbial communities dominated by human-associated bacteria with clear linkages between communities on or in different body sites and those communities found on restroom surfaces. More generally, this work is relevant to the public health field as we show that human associated microbes are commonly found on restroom surfaces suggesting that bacterial pathogens could readily be transmitted between individuals by the touching of surfaces. Furthermore, we demonstrate that we can use high-throughput analyses of bacterial communities to determine sources of bacteria on indoor surfaces, an approach which could be used to track pathogen transmission and test the efficacy of hygiene practices."	ENVO:building	TCAG	Microbial biogeography of public restroom surfaces	NA	0	"Ten surfaces (door handles into and out of the restroom, handles into and out of a restroom stall, faucet handles, soap dispenser, toilet seat, toilet flush handle, floor around the toilet and floor around the sink) in six male and six female restrooms evenly distributed across two buildings on the University of Colorado at Boulder campus were sampled on a single day in November 2010. Surfaces where sampled using sterile, cotton-tipped swabs as described previously?[14],?[15]. As the 12 restrooms were nearly identical in design, we were able to swab the same area at each location between restrooms. In order to characterize tap water communities as a potential source of bacteria, 1 L of faucet water from six of the restrooms (each building having the same water source for each restroom sampled) was collected and filtered through 0.2 ?m bottle top filters (Nalgene, Rochester, NY, USA)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	22132229	NA	NA	NA	Gilberto Flores	CCME	GAZ:United States of America	NA	NA	0	NA	NA	NA	NA	n	NA	Noah Fierer	Flores_restroom_surface_biogeography	pyrosequencing	NA	NA	NA	restroom surface metagenome	1	ENVO:city	Titanium	NA	NA	NA	NA	NA	flores_restroom_surface_biogeography	restroom surface biogeography	256318	CCME	NA	NA	NA	1335	"Genomic DNA was extracted from the swabs and filters using the MO BIO PowerSoil DNA isolation kit following the manufacturer's protocol with the modifications of Fierer?et al.?[14]. A portion of the 16 S rRNA gene spanning the V1_V2 regions was amplified using the primer set (27 F/338R), PCR mixture conditions and thermal cycling conditions described in Fierer?et al.?[15]. PCR amplicons of triplicate reactions for each sample were pooled at approximately equal amounts and pyrosequenced at 454 Life Sciences (Branford, CT, USA) on their GS Junior system. A total of 337,333 high-quality partial 16 S rRNA gene sequences were obtained from 101 of the 120 surface samples collected, averaging approximately 3,340 sequences per sample (ranging from 513_6,771) (Table S1) in 4 GS Junior runs, with the best run containing 116,004 high-quality reads. An additional 16,416 sequences (ranging from 2161_5084 per sample) were generated for five of the six water samples collected for source tracking analysis. Each sample was amplified with a unique barcode to enable multiplexing in the GS Junior runs. The barcoded sequencing reads can be separated by data analysis software providing high confidence in assigning sequencing read to each sample. Sequence data generated as part of this study is available upon request by contacting the corresponding author."	NA	NA	y	NA	NA	ENVO:surface	16S rRNA	Porter	NA	V2	CCME	NA	Door out	NA	NA	NA	40.01499	NA	NA	Door handle area out of restroom
PTBF2.489545	GCTCAGTGCAGA	YATGCTGCCTCCCGTAGGAGT	jesse.stombaugh@colorado.edu	NA	restroom surface	Door_handle_area_out_of_restroom	NA	NA	NA	n	12/13/10	restroom surface metagenome	NA	GFL0IUQ	NA	mimarks-survey	NA	NA	NA	NA	0	NA	NA	1624.1	10-Nov	Female	0	NA	NA	NA	NA	NA	NA	NA	NA	NA	NA	"1,swab"	B	NA	Microbial biogeography of public restroom surfaces	NA	NA	-105.27055	y	NA	CCME	NA	NA	NA	"We spend the majority of our lives indoors where we are constantly exposed to bacteria residing on surfaces. However, the diversity of these surface-associated communities is largely unknown. We explored the biogeographical patterns exhibited by bacteria across ten surfaces within each of twelve public restrooms. Using high-throughput barcoded pyrosequencing of the 16S rRNA gene, we identified 19 bacterial phyla across all surfaces. Most sequences belonged to four phyla: Actinobacteria, Bacteriodetes, Firmicutes and Proteobacteria. The communities clustered into three general categories: those found on surfaces associated with toilets, those on the restroom floor, and those found on surfaces routinely touched with hands. On toilet surfaces, gut-associated taxa were more prevalent, suggesting fecal contamination of these surfaces. Floor surfaces were the most diverse of all communities and contained several taxa commonly found in soils. Skin-associated bacteria, especially the Propionibacteriaceae, dominated surfaces routinely touched with our hands. Certain taxa were more common in female than in male restrooms as vagina-associated Lactobacillaceae were widely distributed in female restrooms, likely from urine contamination. Use of the SourceTracker algorithm confirmed many of our taxonomic observations as human skin was the primary source of bacteria on restroom surfaces. Overall, these results demonstrate that restroom surfaces host relatively diverse microbial communities dominated by human-associated bacteria with clear linkages between communities on or in different body sites and those communities found on restroom surfaces. More generally, this work is relevant to the public health field as we show that human associated microbes are commonly found on restroom surfaces suggesting that bacterial pathogens could readily be transmitted between individuals by the touching of surfaces. Furthermore, we demonstrate that we can use high-throughput analyses of bacterial communities to determine sources of bacteria on indoor surfaces, an approach which could be used to track pathogen transmission and test the efficacy of hygiene practices."	ENVO:building	TCAG	Microbial biogeography of public restroom surfaces	NA	0	"Ten surfaces (door handles into and out of the restroom, handles into and out of a restroom stall, faucet handles, soap dispenser, toilet seat, toilet flush handle, floor around the toilet and floor around the sink) in six male and six female restrooms evenly distributed across two buildings on the University of Colorado at Boulder campus were sampled on a single day in November 2010. Surfaces where sampled using sterile, cotton-tipped swabs as described previously?[14],?[15]. As the 12 restrooms were nearly identical in design, we were able to swab the same area at each location between restrooms. In order to characterize tap water communities as a potential source of bacteria, 1 L of faucet water from six of the restrooms (each building having the same water source for each restroom sampled) was collected and filtered through 0.2 ?m bottle top filters (Nalgene, Rochester, NY, USA)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	22132229	NA	NA	NA	Gilberto Flores	CCME	GAZ:United States of America	NA	NA	0	NA	NA	NA	NA	n	NA	Noah Fierer	Flores_restroom_surface_biogeography	pyrosequencing	NA	NA	NA	restroom surface metagenome	2	ENVO:city	Titanium	NA	NA	NA	NA	NA	flores_restroom_surface_biogeography	restroom surface biogeography	256318	CCME	NA	NA	NA	1335	"Genomic DNA was extracted from the swabs and filters using the MO BIO PowerSoil DNA isolation kit following the manufacturer's protocol with the modifications of Fierer?et al.?[14]. A portion of the 16 S rRNA gene spanning the V1_V2 regions was amplified using the primer set (27 F/338R), PCR mixture conditions and thermal cycling conditions described in Fierer?et al.?[15]. PCR amplicons of triplicate reactions for each sample were pooled at approximately equal amounts and pyrosequenced at 454 Life Sciences (Branford, CT, USA) on their GS Junior system. A total of 337,333 high-quality partial 16 S rRNA gene sequences were obtained from 101 of the 120 surface samples collected, averaging approximately 3,340 sequences per sample (ranging from 513_6,771) (Table S1) in 4 GS Junior runs, with the best run containing 116,004 high-quality reads. An additional 16,416 sequences (ranging from 2161_5084 per sample) were generated for five of the six water samples collected for source tracking analysis. Each sample was amplified with a unique barcode to enable multiplexing in the GS Junior runs. The barcoded sequencing reads can be separated by data analysis software providing high confidence in assigning sequencing read to each sample. Sequence data generated as part of this study is available upon request by contacting the corresponding author."	NA	NA	y	NA	NA	ENVO:surface	16S rRNA	Porter	NA	V2	CCME	NA	Door out	NA	NA	NA	40.01499	NA	NA	Door handle area out of restroom
PTBM2.489516	GCGATATATCGC	YATGCTGCCTCCCGTAGGAGT	jesse.stombaugh@colorado.edu	NA	restroom surface	Door_handle_area_out_of_restroom	NA	NA	NA	n	12/13/10	restroom surface metagenome	NA	GFL0IUQ	NA	mimarks-survey	NA	NA	NA	NA	0	NA	NA	1624.1	10-Nov	Male	0	NA	NA	NA	NA	NA	NA	NA	NA	NA	NA	"1,swab"	B	NA	Microbial biogeography of public restroom surfaces	NA	NA	-105.27055	y	NA	CCME	NA	NA	NA	"We spend the majority of our lives indoors where we are constantly exposed to bacteria residing on surfaces. However, the diversity of these surface-associated communities is largely unknown. We explored the biogeographical patterns exhibited by bacteria across ten surfaces within each of twelve public restrooms. Using high-throughput barcoded pyrosequencing of the 16S rRNA gene, we identified 19 bacterial phyla across all surfaces. Most sequences belonged to four phyla: Actinobacteria, Bacteriodetes, Firmicutes and Proteobacteria. The communities clustered into three general categories: those found on surfaces associated with toilets, those on the restroom floor, and those found on surfaces routinely touched with hands. On toilet surfaces, gut-associated taxa were more prevalent, suggesting fecal contamination of these surfaces. Floor surfaces were the most diverse of all communities and contained several taxa commonly found in soils. Skin-associated bacteria, especially the Propionibacteriaceae, dominated surfaces routinely touched with our hands. Certain taxa were more common in female than in male restrooms as vagina-associated Lactobacillaceae were widely distributed in female restrooms, likely from urine contamination. Use of the SourceTracker algorithm confirmed many of our taxonomic observations as human skin was the primary source of bacteria on restroom surfaces. Overall, these results demonstrate that restroom surfaces host relatively diverse microbial communities dominated by human-associated bacteria with clear linkages between communities on or in different body sites and those communities found on restroom surfaces. More generally, this work is relevant to the public health field as we show that human associated microbes are commonly found on restroom surfaces suggesting that bacterial pathogens could readily be transmitted between individuals by the touching of surfaces. Furthermore, we demonstrate that we can use high-throughput analyses of bacterial communities to determine sources of bacteria on indoor surfaces, an approach which could be used to track pathogen transmission and test the efficacy of hygiene practices."	ENVO:building	TCAG	Microbial biogeography of public restroom surfaces	NA	0	"Ten surfaces (door handles into and out of the restroom, handles into and out of a restroom stall, faucet handles, soap dispenser, toilet seat, toilet flush handle, floor around the toilet and floor around the sink) in six male and six female restrooms evenly distributed across two buildings on the University of Colorado at Boulder campus were sampled on a single day in November 2010. Surfaces where sampled using sterile, cotton-tipped swabs as described previously?[14],?[15]. As the 12 restrooms were nearly identical in design, we were able to swab the same area at each location between restrooms. In order to characterize tap water communities as a potential source of bacteria, 1 L of faucet water from six of the restrooms (each building having the same water source for each restroom sampled) was collected and filtered through 0.2 ?m bottle top filters (Nalgene, Rochester, NY, USA)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	22132229	NA	NA	NA	Gilberto Flores	CCME	GAZ:United States of America	NA	NA	0	NA	NA	NA	NA	n	NA	Noah Fierer	Flores_restroom_surface_biogeography	pyrosequencing	NA	NA	NA	restroom surface metagenome	2	ENVO:city	Titanium	NA	NA	NA	NA	NA	flores_restroom_surface_biogeography	restroom surface biogeography	256318	CCME	NA	NA	NA	1335	"Genomic DNA was extracted from the swabs and filters using the MO BIO PowerSoil DNA isolation kit following the manufacturer's protocol with the modifications of Fierer?et al.?[14]. A portion of the 16 S rRNA gene spanning the V1_V2 regions was amplified using the primer set (27 F/338R), PCR mixture conditions and thermal cycling conditions described in Fierer?et al.?[15]. PCR amplicons of triplicate reactions for each sample were pooled at approximately equal amounts and pyrosequenced at 454 Life Sciences (Branford, CT, USA) on their GS Junior system. A total of 337,333 high-quality partial 16 S rRNA gene sequences were obtained from 101 of the 120 surface samples collected, averaging approximately 3,340 sequences per sample (ranging from 513_6,771) (Table S1) in 4 GS Junior runs, with the best run containing 116,004 high-quality reads. An additional 16,416 sequences (ranging from 2161_5084 per sample) were generated for five of the six water samples collected for source tracking analysis. Each sample was amplified with a unique barcode to enable multiplexing in the GS Junior runs. The barcoded sequencing reads can be separated by data analysis software providing high confidence in assigning sequencing read to each sample. Sequence data generated as part of this study is available upon request by contacting the corresponding author."	NA	NA	y	NA	NA	ENVO:surface	16S rRNA	Porter	NA	V2	CCME	NA	Door out	NA	NA	NA	40.01499	NA	NA	Door handle area out of restroom
PTCF2.489540	ATATCGCTACTG	YATGCTGCCTCCCGTAGGAGT	jesse.stombaugh@colorado.edu	NA	restroom surface	Door_handle_area_out_of_restroom	NA	NA	NA	n	12/13/10	restroom surface metagenome	NA	GFL0IUQ	NA	mimarks-survey	NA	NA	NA	NA	0	NA	NA	1624.1	10-Nov	Female	0	NA	NA	NA	NA	NA	NA	NA	NA	NA	NA	"1,swab"	C	NA	Microbial biogeography of public restroom surfaces	NA	NA	-105.27055	y	NA	CCME	NA	NA	NA	"We spend the majority of our lives indoors where we are constantly exposed to bacteria residing on surfaces. However, the diversity of these surface-associated communities is largely unknown. We explored the biogeographical patterns exhibited by bacteria across ten surfaces within each of twelve public restrooms. Using high-throughput barcoded pyrosequencing of the 16S rRNA gene, we identified 19 bacterial phyla across all surfaces. Most sequences belonged to four phyla: Actinobacteria, Bacteriodetes, Firmicutes and Proteobacteria. The communities clustered into three general categories: those found on surfaces associated with toilets, those on the restroom floor, and those found on surfaces routinely touched with hands. On toilet surfaces, gut-associated taxa were more prevalent, suggesting fecal contamination of these surfaces. Floor surfaces were the most diverse of all communities and contained several taxa commonly found in soils. Skin-associated bacteria, especially the Propionibacteriaceae, dominated surfaces routinely touched with our hands. Certain taxa were more common in female than in male restrooms as vagina-associated Lactobacillaceae were widely distributed in female restrooms, likely from urine contamination. Use of the SourceTracker algorithm confirmed many of our taxonomic observations as human skin was the primary source of bacteria on restroom surfaces. Overall, these results demonstrate that restroom surfaces host relatively diverse microbial communities dominated by human-associated bacteria with clear linkages between communities on or in different body sites and those communities found on restroom surfaces. More generally, this work is relevant to the public health field as we show that human associated microbes are commonly found on restroom surfaces suggesting that bacterial pathogens could readily be transmitted between individuals by the touching of surfaces. Furthermore, we demonstrate that we can use high-throughput analyses of bacterial communities to determine sources of bacteria on indoor surfaces, an approach which could be used to track pathogen transmission and test the efficacy of hygiene practices."	ENVO:building	TCAG	Microbial biogeography of public restroom surfaces	NA	0	"Ten surfaces (door handles into and out of the restroom, handles into and out of a restroom stall, faucet handles, soap dispenser, toilet seat, toilet flush handle, floor around the toilet and floor around the sink) in six male and six female restrooms evenly distributed across two buildings on the University of Colorado at Boulder campus were sampled on a single day in November 2010. Surfaces where sampled using sterile, cotton-tipped swabs as described previously?[14],?[15]. As the 12 restrooms were nearly identical in design, we were able to swab the same area at each location between restrooms. In order to characterize tap water communities as a potential source of bacteria, 1 L of faucet water from six of the restrooms (each building having the same water source for each restroom sampled) was collected and filtered through 0.2 ?m bottle top filters (Nalgene, Rochester, NY, USA)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	22132229	NA	NA	NA	Gilberto Flores	CCME	GAZ:United States of America	NA	NA	0	NA	NA	NA	NA	n	NA	Noah Fierer	Flores_restroom_surface_biogeography	pyrosequencing	NA	NA	NA	restroom surface metagenome	3	ENVO:city	Titanium	NA	NA	NA	NA	NA	flores_restroom_surface_biogeography	restroom surface biogeography	256318	CCME	NA	NA	NA	1335	"Genomic DNA was extracted from the swabs and filters using the MO BIO PowerSoil DNA isolation kit following the manufacturer's protocol with the modifications of Fierer?et al.?[14]. A portion of the 16 S rRNA gene spanning the V1_V2 regions was amplified using the primer set (27 F/338R), PCR mixture conditions and thermal cycling conditions described in Fierer?et al.?[15]. PCR amplicons of triplicate reactions for each sample were pooled at approximately equal amounts and pyrosequenced at 454 Life Sciences (Branford, CT, USA) on their GS Junior system. A total of 337,333 high-quality partial 16 S rRNA gene sequences were obtained from 101 of the 120 surface samples collected, averaging approximately 3,340 sequences per sample (ranging from 513_6,771) (Table S1) in 4 GS Junior runs, with the best run containing 116,004 high-quality reads. An additional 16,416 sequences (ranging from 2161_5084 per sample) were generated for five of the six water samples collected for source tracking analysis. Each sample was amplified with a unique barcode to enable multiplexing in the GS Junior runs. The barcoded sequencing reads can be separated by data analysis software providing high confidence in assigning sequencing read to each sample. Sequence data generated as part of this study is available upon request by contacting the corresponding author."	NA	NA	y	NA	NA	ENVO:surface	16S rRNA	Porter	NA	V2	CCME	NA	Door out	NA	NA	NA	40.01499	NA	NA	Door handle area out of restroom
PTCM2.489469	AGTTCAGACGCT	YATGCTGCCTCCCGTAGGAGT	jesse.stombaugh@colorado.edu	NA	restroom surface	Door_handle_area_out_of_restroom	NA	NA	NA	n	12/13/10	restroom surface metagenome	NA	GFL0IUQ	NA	mimarks-survey	NA	NA	NA	NA	0	NA	NA	1624.1	10-Nov	Male	0	NA	NA	NA	NA	NA	NA	NA	NA	NA	NA	"1,swab"	C	NA	Microbial biogeography of public restroom surfaces	NA	NA	-105.27055	y	NA	CCME	NA	NA	NA	"We spend the majority of our lives indoors where we are constantly exposed to bacteria residing on surfaces. However, the diversity of these surface-associated communities is largely unknown. We explored the biogeographical patterns exhibited by bacteria across ten surfaces within each of twelve public restrooms. Using high-throughput barcoded pyrosequencing of the 16S rRNA gene, we identified 19 bacterial phyla across all surfaces. Most sequences belonged to four phyla: Actinobacteria, Bacteriodetes, Firmicutes and Proteobacteria. The communities clustered into three general categories: those found on surfaces associated with toilets, those on the restroom floor, and those found on surfaces routinely touched with hands. On toilet surfaces, gut-associated taxa were more prevalent, suggesting fecal contamination of these surfaces. Floor surfaces were the most diverse of all communities and contained several taxa commonly found in soils. Skin-associated bacteria, especially the Propionibacteriaceae, dominated surfaces routinely touched with our hands. Certain taxa were more common in female than in male restrooms as vagina-associated Lactobacillaceae were widely distributed in female restrooms, likely from urine contamination. Use of the SourceTracker algorithm confirmed many of our taxonomic observations as human skin was the primary source of bacteria on restroom surfaces. Overall, these results demonstrate that restroom surfaces host relatively diverse microbial communities dominated by human-associated bacteria with clear linkages between communities on or in different body sites and those communities found on restroom surfaces. More generally, this work is relevant to the public health field as we show that human associated microbes are commonly found on restroom surfaces suggesting that bacterial pathogens could readily be transmitted between individuals by the touching of surfaces. Furthermore, we demonstrate that we can use high-throughput analyses of bacterial communities to determine sources of bacteria on indoor surfaces, an approach which could be used to track pathogen transmission and test the efficacy of hygiene practices."	ENVO:building	TCAG	Microbial biogeography of public restroom surfaces	NA	0	"Ten surfaces (door handles into and out of the restroom, handles into and out of a restroom stall, faucet handles, soap dispenser, toilet seat, toilet flush handle, floor around the toilet and floor around the sink) in six male and six female restrooms evenly distributed across two buildings on the University of Colorado at Boulder campus were sampled on a single day in November 2010. Surfaces where sampled using sterile, cotton-tipped swabs as described previously?[14],?[15]. As the 12 restrooms were nearly identical in design, we were able to swab the same area at each location between restrooms. In order to characterize tap water communities as a potential source of bacteria, 1 L of faucet water from six of the restrooms (each building having the same water source for each restroom sampled) was collected and filtered through 0.2 ?m bottle top filters (Nalgene, Rochester, NY, USA)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	22132229	NA	NA	NA	Gilberto Flores	CCME	GAZ:United States of America	NA	NA	0	NA	NA	NA	NA	n	NA	Noah Fierer	Flores_restroom_surface_biogeography	pyrosequencing	NA	NA	NA	restroom surface metagenome	3	ENVO:city	Titanium	NA	NA	NA	NA	NA	flores_restroom_surface_biogeography	restroom surface biogeography	256318	CCME	NA	NA	NA	1335	"Genomic DNA was extracted from the swabs and filters using the MO BIO PowerSoil DNA isolation kit following the manufacturer's protocol with the modifications of Fierer?et al.?[14]. A portion of the 16 S rRNA gene spanning the V1_V2 regions was amplified using the primer set (27 F/338R), PCR mixture conditions and thermal cycling conditions described in Fierer?et al.?[15]. PCR amplicons of triplicate reactions for each sample were pooled at approximately equal amounts and pyrosequenced at 454 Life Sciences (Branford, CT, USA) on their GS Junior system. A total of 337,333 high-quality partial 16 S rRNA gene sequences were obtained from 101 of the 120 surface samples collected, averaging approximately 3,340 sequences per sample (ranging from 513_6,771) (Table S1) in 4 GS Junior runs, with the best run containing 116,004 high-quality reads. An additional 16,416 sequences (ranging from 2161_5084 per sample) were generated for five of the six water samples collected for source tracking analysis. Each sample was amplified with a unique barcode to enable multiplexing in the GS Junior runs. The barcoded sequencing reads can be separated by data analysis software providing high confidence in assigning sequencing read to each sample. Sequence data generated as part of this study is available upon request by contacting the corresponding author."	NA	NA	y	NA	NA	ENVO:surface	16S rRNA	Porter	NA	V2	CCME	NA	Door out	NA	NA	NA	40.01499	NA	NA	Door handle area out of restroom
EKAF3.489546	GACAGGAGATAG	YATGCTGCCTCCCGTAGGAGT	jesse.stombaugh@colorado.edu	NA	restroom surface	Door_lock_area_into_stall	NA	NA	NA	n	12/13/10	restroom surface metagenome	NA	GFL0IUQ	NA	mimarks-survey	NA	NA	NA	NA	0	NA	NA	1624.1	10-Nov	Female	0	NA	NA	NA	NA	NA	NA	NA	NA	NA	NA	"1,swab"	A	NA	Microbial biogeography of public restroom surfaces	NA	NA	-105.27055	y	NA	CCME	NA	NA	NA	"We spend the majority of our lives indoors where we are constantly exposed to bacteria residing on surfaces. However, the diversity of these surface-associated communities is largely unknown. We explored the biogeographical patterns exhibited by bacteria across ten surfaces within each of twelve public restrooms. Using high-throughput barcoded pyrosequencing of the 16S rRNA gene, we identified 19 bacterial phyla across all surfaces. Most sequences belonged to four phyla: Actinobacteria, Bacteriodetes, Firmicutes and Proteobacteria. The communities clustered into three general categories: those found on surfaces associated with toilets, those on the restroom floor, and those found on surfaces routinely touched with hands. On toilet surfaces, gut-associated taxa were more prevalent, suggesting fecal contamination of these surfaces. Floor surfaces were the most diverse of all communities and contained several taxa commonly found in soils. Skin-associated bacteria, especially the Propionibacteriaceae, dominated surfaces routinely touched with our hands. Certain taxa were more common in female than in male restrooms as vagina-associated Lactobacillaceae were widely distributed in female restrooms, likely from urine contamination. Use of the SourceTracker algorithm confirmed many of our taxonomic observations as human skin was the primary source of bacteria on restroom surfaces. Overall, these results demonstrate that restroom surfaces host relatively diverse microbial communities dominated by human-associated bacteria with clear linkages between communities on or in different body sites and those communities found on restroom surfaces. More generally, this work is relevant to the public health field as we show that human associated microbes are commonly found on restroom surfaces suggesting that bacterial pathogens could readily be transmitted between individuals by the touching of surfaces. Furthermore, we demonstrate that we can use high-throughput analyses of bacterial communities to determine sources of bacteria on indoor surfaces, an approach which could be used to track pathogen transmission and test the efficacy of hygiene practices."	ENVO:building	TCAG	Microbial biogeography of public restroom surfaces	NA	0	"Ten surfaces (door handles into and out of the restroom, handles into and out of a restroom stall, faucet handles, soap dispenser, toilet seat, toilet flush handle, floor around the toilet and floor around the sink) in six male and six female restrooms evenly distributed across two buildings on the University of Colorado at Boulder campus were sampled on a single day in November 2010. Surfaces where sampled using sterile, cotton-tipped swabs as described previously?[14],?[15]. As the 12 restrooms were nearly identical in design, we were able to swab the same area at each location between restrooms. In order to characterize tap water communities as a potential source of bacteria, 1 L of faucet water from six of the restrooms (each building having the same water source for each restroom sampled) was collected and filtered through 0.2 ?m bottle top filters (Nalgene, Rochester, NY, USA)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	22132229	NA	NA	NA	Gilberto Flores	CCME	GAZ:United States of America	NA	NA	0	NA	NA	NA	NA	n	NA	Noah Fierer	Flores_restroom_surface_biogeography	pyrosequencing	NA	NA	NA	restroom surface metagenome	1	ENVO:city	Titanium	NA	NA	NA	NA	NA	flores_restroom_surface_biogeography	restroom surface biogeography	256318	CCME	NA	NA	NA	1335	"Genomic DNA was extracted from the swabs and filters using the MO BIO PowerSoil DNA isolation kit following the manufacturer's protocol with the modifications of Fierer?et al.?[14]. A portion of the 16 S rRNA gene spanning the V1_V2 regions was amplified using the primer set (27 F/338R), PCR mixture conditions and thermal cycling conditions described in Fierer?et al.?[15]. PCR amplicons of triplicate reactions for each sample were pooled at approximately equal amounts and pyrosequenced at 454 Life Sciences (Branford, CT, USA) on their GS Junior system. A total of 337,333 high-quality partial 16 S rRNA gene sequences were obtained from 101 of the 120 surface samples collected, averaging approximately 3,340 sequences per sample (ranging from 513_6,771) (Table S1) in 4 GS Junior runs, with the best run containing 116,004 high-quality reads. An additional 16,416 sequences (ranging from 2161_5084 per sample) were generated for five of the six water samples collected for source tracking analysis. Each sample was amplified with a unique barcode to enable multiplexing in the GS Junior runs. The barcoded sequencing reads can be separated by data analysis software providing high confidence in assigning sequencing read to each sample. Sequence data generated as part of this study is available upon request by contacting the corresponding author."	NA	NA	y	NA	NA	ENVO:surface	16S rRNA	Ekeley	NA	V2	CCME	NA	Stall in	NA	NA	NA	40.01499	NA	NA	Door lock area into stall
EKBM3.489450	GACTAACGTCAC	YATGCTGCCTCCCGTAGGAGT	jesse.stombaugh@colorado.edu	NA	restroom surface	Door_lock_area_into_stall	NA	NA	NA	n	12/13/10	restroom surface metagenome	NA	GFL0IUQ	NA	mimarks-survey	NA	NA	NA	NA	0	NA	NA	1624.1	10-Nov	Male	0	NA	NA	NA	NA	NA	NA	NA	NA	NA	NA	"1,swab"	B	NA	Microbial biogeography of public restroom surfaces	NA	NA	-105.27055	y	NA	CCME	NA	NA	NA	"We spend the majority of our lives indoors where we are constantly exposed to bacteria residing on surfaces. However, the diversity of these surface-associated communities is largely unknown. We explored the biogeographical patterns exhibited by bacteria across ten surfaces within each of twelve public restrooms. Using high-throughput barcoded pyrosequencing of the 16S rRNA gene, we identified 19 bacterial phyla across all surfaces. Most sequences belonged to four phyla: Actinobacteria, Bacteriodetes, Firmicutes and Proteobacteria. The communities clustered into three general categories: those found on surfaces associated with toilets, those on the restroom floor, and those found on surfaces routinely touched with hands. On toilet surfaces, gut-associated taxa were more prevalent, suggesting fecal contamination of these surfaces. Floor surfaces were the most diverse of all communities and contained several taxa commonly found in soils. Skin-associated bacteria, especially the Propionibacteriaceae, dominated surfaces routinely touched with our hands. Certain taxa were more common in female than in male restrooms as vagina-associated Lactobacillaceae were widely distributed in female restrooms, likely from urine contamination. Use of the SourceTracker algorithm confirmed many of our taxonomic observations as human skin was the primary source of bacteria on restroom surfaces. Overall, these results demonstrate that restroom surfaces host relatively diverse microbial communities dominated by human-associated bacteria with clear linkages between communities on or in different body sites and those communities found on restroom surfaces. More generally, this work is relevant to the public health field as we show that human associated microbes are commonly found on restroom surfaces suggesting that bacterial pathogens could readily be transmitted between individuals by the touching of surfaces. Furthermore, we demonstrate that we can use high-throughput analyses of bacterial communities to determine sources of bacteria on indoor surfaces, an approach which could be used to track pathogen transmission and test the efficacy of hygiene practices."	ENVO:building	TCAG	Microbial biogeography of public restroom surfaces	NA	0	"Ten surfaces (door handles into and out of the restroom, handles into and out of a restroom stall, faucet handles, soap dispenser, toilet seat, toilet flush handle, floor around the toilet and floor around the sink) in six male and six female restrooms evenly distributed across two buildings on the University of Colorado at Boulder campus were sampled on a single day in November 2010. Surfaces where sampled using sterile, cotton-tipped swabs as described previously?[14],?[15]. As the 12 restrooms were nearly identical in design, we were able to swab the same area at each location between restrooms. In order to characterize tap water communities as a potential source of bacteria, 1 L of faucet water from six of the restrooms (each building having the same water source for each restroom sampled) was collected and filtered through 0.2 ?m bottle top filters (Nalgene, Rochester, NY, USA)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	22132229	NA	NA	NA	Gilberto Flores	CCME	GAZ:United States of America	NA	NA	0	NA	NA	NA	NA	n	NA	Noah Fierer	Flores_restroom_surface_biogeography	pyrosequencing	NA	NA	NA	restroom surface metagenome	2	ENVO:city	Titanium	NA	NA	NA	NA	NA	flores_restroom_surface_biogeography	restroom surface biogeography	256318	CCME	NA	NA	NA	1335	"Genomic DNA was extracted from the swabs and filters using the MO BIO PowerSoil DNA isolation kit following the manufacturer's protocol with the modifications of Fierer?et al.?[14]. A portion of the 16 S rRNA gene spanning the V1_V2 regions was amplified using the primer set (27 F/338R), PCR mixture conditions and thermal cycling conditions described in Fierer?et al.?[15]. PCR amplicons of triplicate reactions for each sample were pooled at approximately equal amounts and pyrosequenced at 454 Life Sciences (Branford, CT, USA) on their GS Junior system. A total of 337,333 high-quality partial 16 S rRNA gene sequences were obtained from 101 of the 120 surface samples collected, averaging approximately 3,340 sequences per sample (ranging from 513_6,771) (Table S1) in 4 GS Junior runs, with the best run containing 116,004 high-quality reads. An additional 16,416 sequences (ranging from 2161_5084 per sample) were generated for five of the six water samples collected for source tracking analysis. Each sample was amplified with a unique barcode to enable multiplexing in the GS Junior runs. The barcoded sequencing reads can be separated by data analysis software providing high confidence in assigning sequencing read to each sample. Sequence data generated as part of this study is available upon request by contacting the corresponding author."	NA	NA	y	NA	NA	ENVO:surface	16S rRNA	Ekeley	NA	V2	CCME	NA	Stall in	NA	NA	NA	40.01499	NA	NA	Door lock area into stall
PTAM3.489461	GCACGACAACAC	YATGCTGCCTCCCGTAGGAGT	jesse.stombaugh@colorado.edu	NA	restroom surface	Door_lock_area_into_stall	NA	NA	NA	n	12/13/10	restroom surface metagenome	NA	GFL0IUQ	NA	mimarks-survey	NA	NA	NA	NA	0	NA	NA	1624.1	10-Nov	Male	0	NA	NA	NA	NA	NA	NA	NA	NA	NA	NA	"1,swab"	A	NA	Microbial biogeography of public restroom surfaces	NA	NA	-105.27055	y	NA	CCME	NA	NA	NA	"We spend the majority of our lives indoors where we are constantly exposed to bacteria residing on surfaces. However, the diversity of these surface-associated communities is largely unknown. We explored the biogeographical patterns exhibited by bacteria across ten surfaces within each of twelve public restrooms. Using high-throughput barcoded pyrosequencing of the 16S rRNA gene, we identified 19 bacterial phyla across all surfaces. Most sequences belonged to four phyla: Actinobacteria, Bacteriodetes, Firmicutes and Proteobacteria. The communities clustered into three general categories: those found on surfaces associated with toilets, those on the restroom floor, and those found on surfaces routinely touched with hands. On toilet surfaces, gut-associated taxa were more prevalent, suggesting fecal contamination of these surfaces. Floor surfaces were the most diverse of all communities and contained several taxa commonly found in soils. Skin-associated bacteria, especially the Propionibacteriaceae, dominated surfaces routinely touched with our hands. Certain taxa were more common in female than in male restrooms as vagina-associated Lactobacillaceae were widely distributed in female restrooms, likely from urine contamination. Use of the SourceTracker algorithm confirmed many of our taxonomic observations as human skin was the primary source of bacteria on restroom surfaces. Overall, these results demonstrate that restroom surfaces host relatively diverse microbial communities dominated by human-associated bacteria with clear linkages between communities on or in different body sites and those communities found on restroom surfaces. More generally, this work is relevant to the public health field as we show that human associated microbes are commonly found on restroom surfaces suggesting that bacterial pathogens could readily be transmitted between individuals by the touching of surfaces. Furthermore, we demonstrate that we can use high-throughput analyses of bacterial communities to determine sources of bacteria on indoor surfaces, an approach which could be used to track pathogen transmission and test the efficacy of hygiene practices."	ENVO:building	TCAG	Microbial biogeography of public restroom surfaces	NA	0	"Ten surfaces (door handles into and out of the restroom, handles into and out of a restroom stall, faucet handles, soap dispenser, toilet seat, toilet flush handle, floor around the toilet and floor around the sink) in six male and six female restrooms evenly distributed across two buildings on the University of Colorado at Boulder campus were sampled on a single day in November 2010. Surfaces where sampled using sterile, cotton-tipped swabs as described previously?[14],?[15]. As the 12 restrooms were nearly identical in design, we were able to swab the same area at each location between restrooms. In order to characterize tap water communities as a potential source of bacteria, 1 L of faucet water from six of the restrooms (each building having the same water source for each restroom sampled) was collected and filtered through 0.2 ?m bottle top filters (Nalgene, Rochester, NY, USA)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	22132229	NA	NA	NA	Gilberto Flores	CCME	GAZ:United States of America	NA	NA	0	NA	NA	NA	NA	n	NA	Noah Fierer	Flores_restroom_surface_biogeography	pyrosequencing	NA	NA	NA	restroom surface metagenome	1	ENVO:city	Titanium	NA	NA	NA	NA	NA	flores_restroom_surface_biogeography	restroom surface biogeography	256318	CCME	NA	NA	NA	1335	"Genomic DNA was extracted from the swabs and filters using the MO BIO PowerSoil DNA isolation kit following the manufacturer's protocol with the modifications of Fierer?et al.?[14]. A portion of the 16 S rRNA gene spanning the V1_V2 regions was amplified using the primer set (27 F/338R), PCR mixture conditions and thermal cycling conditions described in Fierer?et al.?[15]. PCR amplicons of triplicate reactions for each sample were pooled at approximately equal amounts and pyrosequenced at 454 Life Sciences (Branford, CT, USA) on their GS Junior system. A total of 337,333 high-quality partial 16 S rRNA gene sequences were obtained from 101 of the 120 surface samples collected, averaging approximately 3,340 sequences per sample (ranging from 513_6,771) (Table S1) in 4 GS Junior runs, with the best run containing 116,004 high-quality reads. An additional 16,416 sequences (ranging from 2161_5084 per sample) were generated for five of the six water samples collected for source tracking analysis. Each sample was amplified with a unique barcode to enable multiplexing in the GS Junior runs. The barcoded sequencing reads can be separated by data analysis software providing high confidence in assigning sequencing read to each sample. Sequence data generated as part of this study is available upon request by contacting the corresponding author."	NA	NA	y	NA	NA	ENVO:surface	16S rRNA	Porter	NA	V2	CCME	NA	Stall in	NA	NA	NA	40.01499	NA	NA	Door lock area into stall
PTBM3.489559	GCGGATGTGACT	YATGCTGCCTCCCGTAGGAGT	jesse.stombaugh@colorado.edu	NA	restroom surface	Door_lock_area_into_stall	NA	NA	NA	n	12/13/10	restroom surface metagenome	NA	GFL0IUQ	NA	mimarks-survey	NA	NA	NA	NA	0	NA	NA	1624.1	10-Nov	Male	0	NA	NA	NA	NA	NA	NA	NA	NA	NA	NA	"1,swab"	B	NA	Microbial biogeography of public restroom surfaces	NA	NA	-105.27055	y	NA	CCME	NA	NA	NA	"We spend the majority of our lives indoors where we are constantly exposed to bacteria residing on surfaces. However, the diversity of these surface-associated communities is largely unknown. We explored the biogeographical patterns exhibited by bacteria across ten surfaces within each of twelve public restrooms. Using high-throughput barcoded pyrosequencing of the 16S rRNA gene, we identified 19 bacterial phyla across all surfaces. Most sequences belonged to four phyla: Actinobacteria, Bacteriodetes, Firmicutes and Proteobacteria. The communities clustered into three general categories: those found on surfaces associated with toilets, those on the restroom floor, and those found on surfaces routinely touched with hands. On toilet surfaces, gut-associated taxa were more prevalent, suggesting fecal contamination of these surfaces. Floor surfaces were the most diverse of all communities and contained several taxa commonly found in soils. Skin-associated bacteria, especially the Propionibacteriaceae, dominated surfaces routinely touched with our hands. Certain taxa were more common in female than in male restrooms as vagina-associated Lactobacillaceae were widely distributed in female restrooms, likely from urine contamination. Use of the SourceTracker algorithm confirmed many of our taxonomic observations as human skin was the primary source of bacteria on restroom surfaces. Overall, these results demonstrate that restroom surfaces host relatively diverse microbial communities dominated by human-associated bacteria with clear linkages between communities on or in different body sites and those communities found on restroom surfaces. More generally, this work is relevant to the public health field as we show that human associated microbes are commonly found on restroom surfaces suggesting that bacterial pathogens could readily be transmitted between individuals by the touching of surfaces. Furthermore, we demonstrate that we can use high-throughput analyses of bacterial communities to determine sources of bacteria on indoor surfaces, an approach which could be used to track pathogen transmission and test the efficacy of hygiene practices."	ENVO:building	TCAG	Microbial biogeography of public restroom surfaces	NA	0	"Ten surfaces (door handles into and out of the restroom, handles into and out of a restroom stall, faucet handles, soap dispenser, toilet seat, toilet flush handle, floor around the toilet and floor around the sink) in six male and six female restrooms evenly distributed across two buildings on the University of Colorado at Boulder campus were sampled on a single day in November 2010. Surfaces where sampled using sterile, cotton-tipped swabs as described previously?[14],?[15]. As the 12 restrooms were nearly identical in design, we were able to swab the same area at each location between restrooms. In order to characterize tap water communities as a potential source of bacteria, 1 L of faucet water from six of the restrooms (each building having the same water source for each restroom sampled) was collected and filtered through 0.2 ?m bottle top filters (Nalgene, Rochester, NY, USA)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	22132229	NA	NA	NA	Gilberto Flores	CCME	GAZ:United States of America	NA	NA	0	NA	NA	NA	NA	n	NA	Noah Fierer	Flores_restroom_surface_biogeography	pyrosequencing	NA	NA	NA	restroom surface metagenome	2	ENVO:city	Titanium	NA	NA	NA	NA	NA	flores_restroom_surface_biogeography	restroom surface biogeography	256318	CCME	NA	NA	NA	1335	"Genomic DNA was extracted from the swabs and filters using the MO BIO PowerSoil DNA isolation kit following the manufacturer's protocol with the modifications of Fierer?et al.?[14]. A portion of the 16 S rRNA gene spanning the V1_V2 regions was amplified using the primer set (27 F/338R), PCR mixture conditions and thermal cycling conditions described in Fierer?et al.?[15]. PCR amplicons of triplicate reactions for each sample were pooled at approximately equal amounts and pyrosequenced at 454 Life Sciences (Branford, CT, USA) on their GS Junior system. A total of 337,333 high-quality partial 16 S rRNA gene sequences were obtained from 101 of the 120 surface samples collected, averaging approximately 3,340 sequences per sample (ranging from 513_6,771) (Table S1) in 4 GS Junior runs, with the best run containing 116,004 high-quality reads. An additional 16,416 sequences (ranging from 2161_5084 per sample) were generated for five of the six water samples collected for source tracking analysis. Each sample was amplified with a unique barcode to enable multiplexing in the GS Junior runs. The barcoded sequencing reads can be separated by data analysis software providing high confidence in assigning sequencing read to each sample. Sequence data generated as part of this study is available upon request by contacting the corresponding author."	NA	NA	y	NA	NA	ENVO:surface	16S rRNA	Porter	NA	V2	CCME	NA	Stall in	NA	NA	NA	40.01499	NA	NA	Door lock area into stall
PTCF3.489529	ATATGCCAGTGC	YATGCTGCCTCCCGTAGGAGT	jesse.stombaugh@colorado.edu	NA	restroom surface	Door_lock_area_into_stall	NA	NA	NA	n	12/13/10	restroom surface metagenome	NA	GFL0IUQ	NA	mimarks-survey	NA	NA	NA	NA	0	NA	NA	1624.1	10-Nov	Female	0	NA	NA	NA	NA	NA	NA	NA	NA	NA	NA	"1,swab"	C	NA	Microbial biogeography of public restroom surfaces	NA	NA	-105.27055	y	NA	CCME	NA	NA	NA	"We spend the majority of our lives indoors where we are constantly exposed to bacteria residing on surfaces. However, the diversity of these surface-associated communities is largely unknown. We explored the biogeographical patterns exhibited by bacteria across ten surfaces within each of twelve public restrooms. Using high-throughput barcoded pyrosequencing of the 16S rRNA gene, we identified 19 bacterial phyla across all surfaces. Most sequences belonged to four phyla: Actinobacteria, Bacteriodetes, Firmicutes and Proteobacteria. The communities clustered into three general categories: those found on surfaces associated with toilets, those on the restroom floor, and those found on surfaces routinely touched with hands. On toilet surfaces, gut-associated taxa were more prevalent, suggesting fecal contamination of these surfaces. Floor surfaces were the most diverse of all communities and contained several taxa commonly found in soils. Skin-associated bacteria, especially the Propionibacteriaceae, dominated surfaces routinely touched with our hands. Certain taxa were more common in female than in male restrooms as vagina-associated Lactobacillaceae were widely distributed in female restrooms, likely from urine contamination. Use of the SourceTracker algorithm confirmed many of our taxonomic observations as human skin was the primary source of bacteria on restroom surfaces. Overall, these results demonstrate that restroom surfaces host relatively diverse microbial communities dominated by human-associated bacteria with clear linkages between communities on or in different body sites and those communities found on restroom surfaces. More generally, this work is relevant to the public health field as we show that human associated microbes are commonly found on restroom surfaces suggesting that bacterial pathogens could readily be transmitted between individuals by the touching of surfaces. Furthermore, we demonstrate that we can use high-throughput analyses of bacterial communities to determine sources of bacteria on indoor surfaces, an approach which could be used to track pathogen transmission and test the efficacy of hygiene practices."	ENVO:building	TCAG	Microbial biogeography of public restroom surfaces	NA	0	"Ten surfaces (door handles into and out of the restroom, handles into and out of a restroom stall, faucet handles, soap dispenser, toilet seat, toilet flush handle, floor around the toilet and floor around the sink) in six male and six female restrooms evenly distributed across two buildings on the University of Colorado at Boulder campus were sampled on a single day in November 2010. Surfaces where sampled using sterile, cotton-tipped swabs as described previously?[14],?[15]. As the 12 restrooms were nearly identical in design, we were able to swab the same area at each location between restrooms. In order to characterize tap water communities as a potential source of bacteria, 1 L of faucet water from six of the restrooms (each building having the same water source for each restroom sampled) was collected and filtered through 0.2 ?m bottle top filters (Nalgene, Rochester, NY, USA)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	22132229	NA	NA	NA	Gilberto Flores	CCME	GAZ:United States of America	NA	NA	0	NA	NA	NA	NA	n	NA	Noah Fierer	Flores_restroom_surface_biogeography	pyrosequencing	NA	NA	NA	restroom surface metagenome	3	ENVO:city	Titanium	NA	NA	NA	NA	NA	flores_restroom_surface_biogeography	restroom surface biogeography	256318	CCME	NA	NA	NA	1335	"Genomic DNA was extracted from the swabs and filters using the MO BIO PowerSoil DNA isolation kit following the manufacturer's protocol with the modifications of Fierer?et al.?[14]. A portion of the 16 S rRNA gene spanning the V1_V2 regions was amplified using the primer set (27 F/338R), PCR mixture conditions and thermal cycling conditions described in Fierer?et al.?[15]. PCR amplicons of triplicate reactions for each sample were pooled at approximately equal amounts and pyrosequenced at 454 Life Sciences (Branford, CT, USA) on their GS Junior system. A total of 337,333 high-quality partial 16 S rRNA gene sequences were obtained from 101 of the 120 surface samples collected, averaging approximately 3,340 sequences per sample (ranging from 513_6,771) (Table S1) in 4 GS Junior runs, with the best run containing 116,004 high-quality reads. An additional 16,416 sequences (ranging from 2161_5084 per sample) were generated for five of the six water samples collected for source tracking analysis. Each sample was amplified with a unique barcode to enable multiplexing in the GS Junior runs. The barcoded sequencing reads can be separated by data analysis software providing high confidence in assigning sequencing read to each sample. Sequence data generated as part of this study is available upon request by contacting the corresponding author."	NA	NA	y	NA	NA	ENVO:surface	16S rRNA	Porter	NA	V2	CCME	NA	Stall in	NA	NA	NA	40.01499	NA	NA	Door lock area into stall
PTCM3.489508	AGTTCTACGTCA	YATGCTGCCTCCCGTAGGAGT	jesse.stombaugh@colorado.edu	NA	restroom surface	Door_lock_area_into_stall	NA	NA	NA	n	12/13/10	restroom surface metagenome	NA	GFL0IUQ	NA	mimarks-survey	NA	NA	NA	NA	0	NA	NA	1624.1	10-Nov	Male	0	NA	NA	NA	NA	NA	NA	NA	NA	NA	NA	"1,swab"	C	NA	Microbial biogeography of public restroom surfaces	NA	NA	-105.27055	y	NA	CCME	NA	NA	NA	"We spend the majority of our lives indoors where we are constantly exposed to bacteria residing on surfaces. However, the diversity of these surface-associated communities is largely unknown. We explored the biogeographical patterns exhibited by bacteria across ten surfaces within each of twelve public restrooms. Using high-throughput barcoded pyrosequencing of the 16S rRNA gene, we identified 19 bacterial phyla across all surfaces. Most sequences belonged to four phyla: Actinobacteria, Bacteriodetes, Firmicutes and Proteobacteria. The communities clustered into three general categories: those found on surfaces associated with toilets, those on the restroom floor, and those found on surfaces routinely touched with hands. On toilet surfaces, gut-associated taxa were more prevalent, suggesting fecal contamination of these surfaces. Floor surfaces were the most diverse of all communities and contained several taxa commonly found in soils. Skin-associated bacteria, especially the Propionibacteriaceae, dominated surfaces routinely touched with our hands. Certain taxa were more common in female than in male restrooms as vagina-associated Lactobacillaceae were widely distributed in female restrooms, likely from urine contamination. Use of the SourceTracker algorithm confirmed many of our taxonomic observations as human skin was the primary source of bacteria on restroom surfaces. Overall, these results demonstrate that restroom surfaces host relatively diverse microbial communities dominated by human-associated bacteria with clear linkages between communities on or in different body sites and those communities found on restroom surfaces. More generally, this work is relevant to the public health field as we show that human associated microbes are commonly found on restroom surfaces suggesting that bacterial pathogens could readily be transmitted between individuals by the touching of surfaces. Furthermore, we demonstrate that we can use high-throughput analyses of bacterial communities to determine sources of bacteria on indoor surfaces, an approach which could be used to track pathogen transmission and test the efficacy of hygiene practices."	ENVO:building	TCAG	Microbial biogeography of public restroom surfaces	NA	0	"Ten surfaces (door handles into and out of the restroom, handles into and out of a restroom stall, faucet handles, soap dispenser, toilet seat, toilet flush handle, floor around the toilet and floor around the sink) in six male and six female restrooms evenly distributed across two buildings on the University of Colorado at Boulder campus were sampled on a single day in November 2010. Surfaces where sampled using sterile, cotton-tipped swabs as described previously?[14],?[15]. As the 12 restrooms were nearly identical in design, we were able to swab the same area at each location between restrooms. In order to characterize tap water communities as a potential source of bacteria, 1 L of faucet water from six of the restrooms (each building having the same water source for each restroom sampled) was collected and filtered through 0.2 ?m bottle top filters (Nalgene, Rochester, NY, USA)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	22132229	NA	NA	NA	Gilberto Flores	CCME	GAZ:United States of America	NA	NA	0	NA	NA	NA	NA	n	NA	Noah Fierer	Flores_restroom_surface_biogeography	pyrosequencing	NA	NA	NA	restroom surface metagenome	3	ENVO:city	Titanium	NA	NA	NA	NA	NA	flores_restroom_surface_biogeography	restroom surface biogeography	256318	CCME	NA	NA	NA	1335	"Genomic DNA was extracted from the swabs and filters using the MO BIO PowerSoil DNA isolation kit following the manufacturer's protocol with the modifications of Fierer?et al.?[14]. A portion of the 16 S rRNA gene spanning the V1_V2 regions was amplified using the primer set (27 F/338R), PCR mixture conditions and thermal cycling conditions described in Fierer?et al.?[15]. PCR amplicons of triplicate reactions for each sample were pooled at approximately equal amounts and pyrosequenced at 454 Life Sciences (Branford, CT, USA) on their GS Junior system. A total of 337,333 high-quality partial 16 S rRNA gene sequences were obtained from 101 of the 120 surface samples collected, averaging approximately 3,340 sequences per sample (ranging from 513_6,771) (Table S1) in 4 GS Junior runs, with the best run containing 116,004 high-quality reads. An additional 16,416 sequences (ranging from 2161_5084 per sample) were generated for five of the six water samples collected for source tracking analysis. Each sample was amplified with a unique barcode to enable multiplexing in the GS Junior runs. The barcoded sequencing reads can be separated by data analysis software providing high confidence in assigning sequencing read to each sample. Sequence data generated as part of this study is available upon request by contacting the corresponding author."	NA	NA	y	NA	NA	ENVO:surface	16S rRNA	Porter	NA	V2	CCME	NA	Stall in	NA	NA	NA	40.01499	NA	NA	Door lock area into stall
EKAF4.489521	GACAGTTACTGC	YATGCTGCCTCCCGTAGGAGT	jesse.stombaugh@colorado.edu	NA	restroom surface	Door_lock_area_out_of_stall	NA	NA	NA	n	12/13/10	restroom surface metagenome	NA	GFL0IUQ	NA	mimarks-survey	NA	NA	NA	NA	0	NA	NA	1624.1	10-Nov	Female	0	NA	NA	NA	NA	NA	NA	NA	NA	NA	NA	"1,swab"	A	NA	Microbial biogeography of public restroom surfaces	NA	NA	-105.27055	y	NA	CCME	NA	NA	NA	"We spend the majority of our lives indoors where we are constantly exposed to bacteria residing on surfaces. However, the diversity of these surface-associated communities is largely unknown. We explored the biogeographical patterns exhibited by bacteria across ten surfaces within each of twelve public restrooms. Using high-throughput barcoded pyrosequencing of the 16S rRNA gene, we identified 19 bacterial phyla across all surfaces. Most sequences belonged to four phyla: Actinobacteria, Bacteriodetes, Firmicutes and Proteobacteria. The communities clustered into three general categories: those found on surfaces associated with toilets, those on the restroom floor, and those found on surfaces routinely touched with hands. On toilet surfaces, gut-associated taxa were more prevalent, suggesting fecal contamination of these surfaces. Floor surfaces were the most diverse of all communities and contained several taxa commonly found in soils. Skin-associated bacteria, especially the Propionibacteriaceae, dominated surfaces routinely touched with our hands. Certain taxa were more common in female than in male restrooms as vagina-associated Lactobacillaceae were widely distributed in female restrooms, likely from urine contamination. Use of the SourceTracker algorithm confirmed many of our taxonomic observations as human skin was the primary source of bacteria on restroom surfaces. Overall, these results demonstrate that restroom surfaces host relatively diverse microbial communities dominated by human-associated bacteria with clear linkages between communities on or in different body sites and those communities found on restroom surfaces. More generally, this work is relevant to the public health field as we show that human associated microbes are commonly found on restroom surfaces suggesting that bacterial pathogens could readily be transmitted between individuals by the touching of surfaces. Furthermore, we demonstrate that we can use high-throughput analyses of bacterial communities to determine sources of bacteria on indoor surfaces, an approach which could be used to track pathogen transmission and test the efficacy of hygiene practices."	ENVO:building	TCAG	Microbial biogeography of public restroom surfaces	NA	0	"Ten surfaces (door handles into and out of the restroom, handles into and out of a restroom stall, faucet handles, soap dispenser, toilet seat, toilet flush handle, floor around the toilet and floor around the sink) in six male and six female restrooms evenly distributed across two buildings on the University of Colorado at Boulder campus were sampled on a single day in November 2010. Surfaces where sampled using sterile, cotton-tipped swabs as described previously?[14],?[15]. As the 12 restrooms were nearly identical in design, we were able to swab the same area at each location between restrooms. In order to characterize tap water communities as a potential source of bacteria, 1 L of faucet water from six of the restrooms (each building having the same water source for each restroom sampled) was collected and filtered through 0.2 ?m bottle top filters (Nalgene, Rochester, NY, USA)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	22132229	NA	NA	NA	Gilberto Flores	CCME	GAZ:United States of America	NA	NA	0	NA	NA	NA	NA	n	NA	Noah Fierer	Flores_restroom_surface_biogeography	pyrosequencing	NA	NA	NA	restroom surface metagenome	1	ENVO:city	Titanium	NA	NA	NA	NA	NA	flores_restroom_surface_biogeography	restroom surface biogeography	256318	CCME	NA	NA	NA	1335	"Genomic DNA was extracted from the swabs and filters using the MO BIO PowerSoil DNA isolation kit following the manufacturer's protocol with the modifications of Fierer?et al.?[14]. A portion of the 16 S rRNA gene spanning the V1_V2 regions was amplified using the primer set (27 F/338R), PCR mixture conditions and thermal cycling conditions described in Fierer?et al.?[15]. PCR amplicons of triplicate reactions for each sample were pooled at approximately equal amounts and pyrosequenced at 454 Life Sciences (Branford, CT, USA) on their GS Junior system. A total of 337,333 high-quality partial 16 S rRNA gene sequences were obtained from 101 of the 120 surface samples collected, averaging approximately 3,340 sequences per sample (ranging from 513_6,771) (Table S1) in 4 GS Junior runs, with the best run containing 116,004 high-quality reads. An additional 16,416 sequences (ranging from 2161_5084 per sample) were generated for five of the six water samples collected for source tracking analysis. Each sample was amplified with a unique barcode to enable multiplexing in the GS Junior runs. The barcoded sequencing reads can be separated by data analysis software providing high confidence in assigning sequencing read to each sample. Sequence data generated as part of this study is available upon request by contacting the corresponding author."	NA	NA	y	NA	NA	ENVO:surface	16S rRNA	Ekeley	NA	V2	CCME	NA	Stall out	NA	NA	NA	40.01499	NA	NA	Door lock area out of stall
EKAM4.489564	CTTGTGTCGATA	YATGCTGCCTCCCGTAGGAGT	jesse.stombaugh@colorado.edu	NA	restroom surface	Door_lock_area_out_of_stall	NA	NA	NA	n	12/13/10	restroom surface metagenome	NA	GFL0IUQ	NA	mimarks-survey	NA	NA	NA	NA	0	NA	NA	1624.1	10-Nov	Male	0	NA	NA	NA	NA	NA	NA	NA	NA	NA	NA	"1,swab"	A	NA	Microbial biogeography of public restroom surfaces	NA	NA	-105.27055	y	NA	CCME	NA	NA	NA	"We spend the majority of our lives indoors where we are constantly exposed to bacteria residing on surfaces. However, the diversity of these surface-associated communities is largely unknown. We explored the biogeographical patterns exhibited by bacteria across ten surfaces within each of twelve public restrooms. Using high-throughput barcoded pyrosequencing of the 16S rRNA gene, we identified 19 bacterial phyla across all surfaces. Most sequences belonged to four phyla: Actinobacteria, Bacteriodetes, Firmicutes and Proteobacteria. The communities clustered into three general categories: those found on surfaces associated with toilets, those on the restroom floor, and those found on surfaces routinely touched with hands. On toilet surfaces, gut-associated taxa were more prevalent, suggesting fecal contamination of these surfaces. Floor surfaces were the most diverse of all communities and contained several taxa commonly found in soils. Skin-associated bacteria, especially the Propionibacteriaceae, dominated surfaces routinely touched with our hands. Certain taxa were more common in female than in male restrooms as vagina-associated Lactobacillaceae were widely distributed in female restrooms, likely from urine contamination. Use of the SourceTracker algorithm confirmed many of our taxonomic observations as human skin was the primary source of bacteria on restroom surfaces. Overall, these results demonstrate that restroom surfaces host relatively diverse microbial communities dominated by human-associated bacteria with clear linkages between communities on or in different body sites and those communities found on restroom surfaces. More generally, this work is relevant to the public health field as we show that human associated microbes are commonly found on restroom surfaces suggesting that bacterial pathogens could readily be transmitted between individuals by the touching of surfaces. Furthermore, we demonstrate that we can use high-throughput analyses of bacterial communities to determine sources of bacteria on indoor surfaces, an approach which could be used to track pathogen transmission and test the efficacy of hygiene practices."	ENVO:building	TCAG	Microbial biogeography of public restroom surfaces	NA	0	"Ten surfaces (door handles into and out of the restroom, handles into and out of a restroom stall, faucet handles, soap dispenser, toilet seat, toilet flush handle, floor around the toilet and floor around the sink) in six male and six female restrooms evenly distributed across two buildings on the University of Colorado at Boulder campus were sampled on a single day in November 2010. Surfaces where sampled using sterile, cotton-tipped swabs as described previously?[14],?[15]. As the 12 restrooms were nearly identical in design, we were able to swab the same area at each location between restrooms. In order to characterize tap water communities as a potential source of bacteria, 1 L of faucet water from six of the restrooms (each building having the same water source for each restroom sampled) was collected and filtered through 0.2 ?m bottle top filters (Nalgene, Rochester, NY, USA)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	22132229	NA	NA	NA	Gilberto Flores	CCME	GAZ:United States of America	NA	NA	0	NA	NA	NA	NA	n	NA	Noah Fierer	Flores_restroom_surface_biogeography	pyrosequencing	NA	NA	NA	restroom surface metagenome	1	ENVO:city	Titanium	NA	NA	NA	NA	NA	flores_restroom_surface_biogeography	restroom surface biogeography	256318	CCME	NA	NA	NA	1335	"Genomic DNA was extracted from the swabs and filters using the MO BIO PowerSoil DNA isolation kit following the manufacturer's protocol with the modifications of Fierer?et al.?[14]. A portion of the 16 S rRNA gene spanning the V1_V2 regions was amplified using the primer set (27 F/338R), PCR mixture conditions and thermal cycling conditions described in Fierer?et al.?[15]. PCR amplicons of triplicate reactions for each sample were pooled at approximately equal amounts and pyrosequenced at 454 Life Sciences (Branford, CT, USA) on their GS Junior system. A total of 337,333 high-quality partial 16 S rRNA gene sequences were obtained from 101 of the 120 surface samples collected, averaging approximately 3,340 sequences per sample (ranging from 513_6,771) (Table S1) in 4 GS Junior runs, with the best run containing 116,004 high-quality reads. An additional 16,416 sequences (ranging from 2161_5084 per sample) were generated for five of the six water samples collected for source tracking analysis. Each sample was amplified with a unique barcode to enable multiplexing in the GS Junior runs. The barcoded sequencing reads can be separated by data analysis software providing high confidence in assigning sequencing read to each sample. Sequence data generated as part of this study is available upon request by contacting the corresponding author."	NA	NA	y	NA	NA	ENVO:surface	16S rRNA	Ekeley	NA	V2	CCME	NA	Stall out	NA	NA	NA	40.01499	NA	NA	Door lock area out of stall
EKBF4.489491	GAGAGCTCTACG	YATGCTGCCTCCCGTAGGAGT	jesse.stombaugh@colorado.edu	NA	restroom surface	Door_lock_area_out_of_stall	NA	NA	NA	n	12/13/10	restroom surface metagenome	NA	GFL0IUQ	NA	mimarks-survey	NA	NA	NA	NA	0	NA	NA	1624.1	10-Nov	Female	0	NA	NA	NA	NA	NA	NA	NA	NA	NA	NA	"1,swab"	B	NA	Microbial biogeography of public restroom surfaces	NA	NA	-105.27055	y	NA	CCME	NA	NA	NA	"We spend the majority of our lives indoors where we are constantly exposed to bacteria residing on surfaces. However, the diversity of these surface-associated communities is largely unknown. We explored the biogeographical patterns exhibited by bacteria across ten surfaces within each of twelve public restrooms. Using high-throughput barcoded pyrosequencing of the 16S rRNA gene, we identified 19 bacterial phyla across all surfaces. Most sequences belonged to four phyla: Actinobacteria, Bacteriodetes, Firmicutes and Proteobacteria. The communities clustered into three general categories: those found on surfaces associated with toilets, those on the restroom floor, and those found on surfaces routinely touched with hands. On toilet surfaces, gut-associated taxa were more prevalent, suggesting fecal contamination of these surfaces. Floor surfaces were the most diverse of all communities and contained several taxa commonly found in soils. Skin-associated bacteria, especially the Propionibacteriaceae, dominated surfaces routinely touched with our hands. Certain taxa were more common in female than in male restrooms as vagina-associated Lactobacillaceae were widely distributed in female restrooms, likely from urine contamination. Use of the SourceTracker algorithm confirmed many of our taxonomic observations as human skin was the primary source of bacteria on restroom surfaces. Overall, these results demonstrate that restroom surfaces host relatively diverse microbial communities dominated by human-associated bacteria with clear linkages between communities on or in different body sites and those communities found on restroom surfaces. More generally, this work is relevant to the public health field as we show that human associated microbes are commonly found on restroom surfaces suggesting that bacterial pathogens could readily be transmitted between individuals by the touching of surfaces. Furthermore, we demonstrate that we can use high-throughput analyses of bacterial communities to determine sources of bacteria on indoor surfaces, an approach which could be used to track pathogen transmission and test the efficacy of hygiene practices."	ENVO:building	TCAG	Microbial biogeography of public restroom surfaces	NA	0	"Ten surfaces (door handles into and out of the restroom, handles into and out of a restroom stall, faucet handles, soap dispenser, toilet seat, toilet flush handle, floor around the toilet and floor around the sink) in six male and six female restrooms evenly distributed across two buildings on the University of Colorado at Boulder campus were sampled on a single day in November 2010. Surfaces where sampled using sterile, cotton-tipped swabs as described previously?[14],?[15]. As the 12 restrooms were nearly identical in design, we were able to swab the same area at each location between restrooms. In order to characterize tap water communities as a potential source of bacteria, 1 L of faucet water from six of the restrooms (each building having the same water source for each restroom sampled) was collected and filtered through 0.2 ?m bottle top filters (Nalgene, Rochester, NY, USA)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	22132229	NA	NA	NA	Gilberto Flores	CCME	GAZ:United States of America	NA	NA	0	NA	NA	NA	NA	n	NA	Noah Fierer	Flores_restroom_surface_biogeography	pyrosequencing	NA	NA	NA	restroom surface metagenome	2	ENVO:city	Titanium	NA	NA	NA	NA	NA	flores_restroom_surface_biogeography	restroom surface biogeography	256318	CCME	NA	NA	NA	1335	"Genomic DNA was extracted from the swabs and filters using the MO BIO PowerSoil DNA isolation kit following the manufacturer's protocol with the modifications of Fierer?et al.?[14]. A portion of the 16 S rRNA gene spanning the V1_V2 regions was amplified using the primer set (27 F/338R), PCR mixture conditions and thermal cycling conditions described in Fierer?et al.?[15]. PCR amplicons of triplicate reactions for each sample were pooled at approximately equal amounts and pyrosequenced at 454 Life Sciences (Branford, CT, USA) on their GS Junior system. A total of 337,333 high-quality partial 16 S rRNA gene sequences were obtained from 101 of the 120 surface samples collected, averaging approximately 3,340 sequences per sample (ranging from 513_6,771) (Table S1) in 4 GS Junior runs, with the best run containing 116,004 high-quality reads. An additional 16,416 sequences (ranging from 2161_5084 per sample) were generated for five of the six water samples collected for source tracking analysis. Each sample was amplified with a unique barcode to enable multiplexing in the GS Junior runs. The barcoded sequencing reads can be separated by data analysis software providing high confidence in assigning sequencing read to each sample. Sequence data generated as part of this study is available upon request by contacting the corresponding author."	NA	NA	y	NA	NA	ENVO:surface	16S rRNA	Ekeley	NA	V2	CCME	NA	Stall out	NA	NA	NA	40.01499	NA	NA	Door lock area out of stall
EKBM4.489515	GACTAGACCAGC	YATGCTGCCTCCCGTAGGAGT	jesse.stombaugh@colorado.edu	NA	restroom surface	Door_lock_area_out_of_stall	NA	NA	NA	n	12/13/10	restroom surface metagenome	NA	GFL0IUQ	NA	mimarks-survey	NA	NA	NA	NA	0	NA	NA	1624.1	10-Nov	Male	0	NA	NA	NA	NA	NA	NA	NA	NA	NA	NA	"1,swab"	B	NA	Microbial biogeography of public restroom surfaces	NA	NA	-105.27055	y	NA	CCME	NA	NA	NA	"We spend the majority of our lives indoors where we are constantly exposed to bacteria residing on surfaces. However, the diversity of these surface-associated communities is largely unknown. We explored the biogeographical patterns exhibited by bacteria across ten surfaces within each of twelve public restrooms. Using high-throughput barcoded pyrosequencing of the 16S rRNA gene, we identified 19 bacterial phyla across all surfaces. Most sequences belonged to four phyla: Actinobacteria, Bacteriodetes, Firmicutes and Proteobacteria. The communities clustered into three general categories: those found on surfaces associated with toilets, those on the restroom floor, and those found on surfaces routinely touched with hands. On toilet surfaces, gut-associated taxa were more prevalent, suggesting fecal contamination of these surfaces. Floor surfaces were the most diverse of all communities and contained several taxa commonly found in soils. Skin-associated bacteria, especially the Propionibacteriaceae, dominated surfaces routinely touched with our hands. Certain taxa were more common in female than in male restrooms as vagina-associated Lactobacillaceae were widely distributed in female restrooms, likely from urine contamination. Use of the SourceTracker algorithm confirmed many of our taxonomic observations as human skin was the primary source of bacteria on restroom surfaces. Overall, these results demonstrate that restroom surfaces host relatively diverse microbial communities dominated by human-associated bacteria with clear linkages between communities on or in different body sites and those communities found on restroom surfaces. More generally, this work is relevant to the public health field as we show that human associated microbes are commonly found on restroom surfaces suggesting that bacterial pathogens could readily be transmitted between individuals by the touching of surfaces. Furthermore, we demonstrate that we can use high-throughput analyses of bacterial communities to determine sources of bacteria on indoor surfaces, an approach which could be used to track pathogen transmission and test the efficacy of hygiene practices."	ENVO:building	TCAG	Microbial biogeography of public restroom surfaces	NA	0	"Ten surfaces (door handles into and out of the restroom, handles into and out of a restroom stall, faucet handles, soap dispenser, toilet seat, toilet flush handle, floor around the toilet and floor around the sink) in six male and six female restrooms evenly distributed across two buildings on the University of Colorado at Boulder campus were sampled on a single day in November 2010. Surfaces where sampled using sterile, cotton-tipped swabs as described previously?[14],?[15]. As the 12 restrooms were nearly identical in design, we were able to swab the same area at each location between restrooms. In order to characterize tap water communities as a potential source of bacteria, 1 L of faucet water from six of the restrooms (each building having the same water source for each restroom sampled) was collected and filtered through 0.2 ?m bottle top filters (Nalgene, Rochester, NY, USA)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	22132229	NA	NA	NA	Gilberto Flores	CCME	GAZ:United States of America	NA	NA	0	NA	NA	NA	NA	n	NA	Noah Fierer	Flores_restroom_surface_biogeography	pyrosequencing	NA	NA	NA	restroom surface metagenome	2	ENVO:city	Titanium	NA	NA	NA	NA	NA	flores_restroom_surface_biogeography	restroom surface biogeography	256318	CCME	NA	NA	NA	1335	"Genomic DNA was extracted from the swabs and filters using the MO BIO PowerSoil DNA isolation kit following the manufacturer's protocol with the modifications of Fierer?et al.?[14]. A portion of the 16 S rRNA gene spanning the V1_V2 regions was amplified using the primer set (27 F/338R), PCR mixture conditions and thermal cycling conditions described in Fierer?et al.?[15]. PCR amplicons of triplicate reactions for each sample were pooled at approximately equal amounts and pyrosequenced at 454 Life Sciences (Branford, CT, USA) on their GS Junior system. A total of 337,333 high-quality partial 16 S rRNA gene sequences were obtained from 101 of the 120 surface samples collected, averaging approximately 3,340 sequences per sample (ranging from 513_6,771) (Table S1) in 4 GS Junior runs, with the best run containing 116,004 high-quality reads. An additional 16,416 sequences (ranging from 2161_5084 per sample) were generated for five of the six water samples collected for source tracking analysis. Each sample was amplified with a unique barcode to enable multiplexing in the GS Junior runs. The barcoded sequencing reads can be separated by data analysis software providing high confidence in assigning sequencing read to each sample. Sequence data generated as part of this study is available upon request by contacting the corresponding author."	NA	NA	y	NA	NA	ENVO:surface	16S rRNA	Ekeley	NA	V2	CCME	NA	Stall out	NA	NA	NA	40.01499	NA	NA	Door lock area out of stall
EKCF4.489498	GATCTCATAGGC	YATGCTGCCTCCCGTAGGAGT	jesse.stombaugh@colorado.edu	NA	restroom surface	Door_lock_area_out_of_stall	NA	NA	NA	n	12/13/10	restroom surface metagenome	NA	GFL0IUQ	NA	mimarks-survey	NA	NA	NA	NA	0	NA	NA	1624.1	10-Nov	Female	0	NA	NA	NA	NA	NA	NA	NA	NA	NA	NA	"1,swab"	C	NA	Microbial biogeography of public restroom surfaces	NA	NA	-105.27055	y	NA	CCME	NA	NA	NA	"We spend the majority of our lives indoors where we are constantly exposed to bacteria residing on surfaces. However, the diversity of these surface-associated communities is largely unknown. We explored the biogeographical patterns exhibited by bacteria across ten surfaces within each of twelve public restrooms. Using high-throughput barcoded pyrosequencing of the 16S rRNA gene, we identified 19 bacterial phyla across all surfaces. Most sequences belonged to four phyla: Actinobacteria, Bacteriodetes, Firmicutes and Proteobacteria. The communities clustered into three general categories: those found on surfaces associated with toilets, those on the restroom floor, and those found on surfaces routinely touched with hands. On toilet surfaces, gut-associated taxa were more prevalent, suggesting fecal contamination of these surfaces. Floor surfaces were the most diverse of all communities and contained several taxa commonly found in soils. Skin-associated bacteria, especially the Propionibacteriaceae, dominated surfaces routinely touched with our hands. Certain taxa were more common in female than in male restrooms as vagina-associated Lactobacillaceae were widely distributed in female restrooms, likely from urine contamination. Use of the SourceTracker algorithm confirmed many of our taxonomic observations as human skin was the primary source of bacteria on restroom surfaces. Overall, these results demonstrate that restroom surfaces host relatively diverse microbial communities dominated by human-associated bacteria with clear linkages between communities on or in different body sites and those communities found on restroom surfaces. More generally, this work is relevant to the public health field as we show that human associated microbes are commonly found on restroom surfaces suggesting that bacterial pathogens could readily be transmitted between individuals by the touching of surfaces. Furthermore, we demonstrate that we can use high-throughput analyses of bacterial communities to determine sources of bacteria on indoor surfaces, an approach which could be used to track pathogen transmission and test the efficacy of hygiene practices."	ENVO:building	TCAG	Microbial biogeography of public restroom surfaces	NA	0	"Ten surfaces (door handles into and out of the restroom, handles into and out of a restroom stall, faucet handles, soap dispenser, toilet seat, toilet flush handle, floor around the toilet and floor around the sink) in six male and six female restrooms evenly distributed across two buildings on the University of Colorado at Boulder campus were sampled on a single day in November 2010. Surfaces where sampled using sterile, cotton-tipped swabs as described previously?[14],?[15]. As the 12 restrooms were nearly identical in design, we were able to swab the same area at each location between restrooms. In order to characterize tap water communities as a potential source of bacteria, 1 L of faucet water from six of the restrooms (each building having the same water source for each restroom sampled) was collected and filtered through 0.2 ?m bottle top filters (Nalgene, Rochester, NY, USA)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	22132229	NA	NA	NA	Gilberto Flores	CCME	GAZ:United States of America	NA	NA	0	NA	NA	NA	NA	n	NA	Noah Fierer	Flores_restroom_surface_biogeography	pyrosequencing	NA	NA	NA	restroom surface metagenome	3	ENVO:city	Titanium	NA	NA	NA	NA	NA	flores_restroom_surface_biogeography	restroom surface biogeography	256318	CCME	NA	NA	NA	1335	"Genomic DNA was extracted from the swabs and filters using the MO BIO PowerSoil DNA isolation kit following the manufacturer's protocol with the modifications of Fierer?et al.?[14]. A portion of the 16 S rRNA gene spanning the V1_V2 regions was amplified using the primer set (27 F/338R), PCR mixture conditions and thermal cycling conditions described in Fierer?et al.?[15]. PCR amplicons of triplicate reactions for each sample were pooled at approximately equal amounts and pyrosequenced at 454 Life Sciences (Branford, CT, USA) on their GS Junior system. A total of 337,333 high-quality partial 16 S rRNA gene sequences were obtained from 101 of the 120 surface samples collected, averaging approximately 3,340 sequences per sample (ranging from 513_6,771) (Table S1) in 4 GS Junior runs, with the best run containing 116,004 high-quality reads. An additional 16,416 sequences (ranging from 2161_5084 per sample) were generated for five of the six water samples collected for source tracking analysis. Each sample was amplified with a unique barcode to enable multiplexing in the GS Junior runs. The barcoded sequencing reads can be separated by data analysis software providing high confidence in assigning sequencing read to each sample. Sequence data generated as part of this study is available upon request by contacting the corresponding author."	NA	NA	y	NA	NA	ENVO:surface	16S rRNA	Ekeley	NA	V2	CCME	NA	Stall out	NA	NA	NA	40.01499	NA	NA	Door lock area out of stall
EKCM4.489509	GAGTGGTAGAGA	YATGCTGCCTCCCGTAGGAGT	jesse.stombaugh@colorado.edu	NA	restroom surface	Door_lock_area_out_of_stall	NA	NA	NA	n	12/13/10	restroom surface metagenome	NA	GFL0IUQ	NA	mimarks-survey	NA	NA	NA	NA	0	NA	NA	1624.1	10-Nov	Male	0	NA	NA	NA	NA	NA	NA	NA	NA	NA	NA	"1,swab"	C	NA	Microbial biogeography of public restroom surfaces	NA	NA	-105.27055	y	NA	CCME	NA	NA	NA	"We spend the majority of our lives indoors where we are constantly exposed to bacteria residing on surfaces. However, the diversity of these surface-associated communities is largely unknown. We explored the biogeographical patterns exhibited by bacteria across ten surfaces within each of twelve public restrooms. Using high-throughput barcoded pyrosequencing of the 16S rRNA gene, we identified 19 bacterial phyla across all surfaces. Most sequences belonged to four phyla: Actinobacteria, Bacteriodetes, Firmicutes and Proteobacteria. The communities clustered into three general categories: those found on surfaces associated with toilets, those on the restroom floor, and those found on surfaces routinely touched with hands. On toilet surfaces, gut-associated taxa were more prevalent, suggesting fecal contamination of these surfaces. Floor surfaces were the most diverse of all communities and contained several taxa commonly found in soils. Skin-associated bacteria, especially the Propionibacteriaceae, dominated surfaces routinely touched with our hands. Certain taxa were more common in female than in male restrooms as vagina-associated Lactobacillaceae were widely distributed in female restrooms, likely from urine contamination. Use of the SourceTracker algorithm confirmed many of our taxonomic observations as human skin was the primary source of bacteria on restroom surfaces. Overall, these results demonstrate that restroom surfaces host relatively diverse microbial communities dominated by human-associated bacteria with clear linkages between communities on or in different body sites and those communities found on restroom surfaces. More generally, this work is relevant to the public health field as we show that human associated microbes are commonly found on restroom surfaces suggesting that bacterial pathogens could readily be transmitted between individuals by the touching of surfaces. Furthermore, we demonstrate that we can use high-throughput analyses of bacterial communities to determine sources of bacteria on indoor surfaces, an approach which could be used to track pathogen transmission and test the efficacy of hygiene practices."	ENVO:building	TCAG	Microbial biogeography of public restroom surfaces	NA	0	"Ten surfaces (door handles into and out of the restroom, handles into and out of a restroom stall, faucet handles, soap dispenser, toilet seat, toilet flush handle, floor around the toilet and floor around the sink) in six male and six female restrooms evenly distributed across two buildings on the University of Colorado at Boulder campus were sampled on a single day in November 2010. Surfaces where sampled using sterile, cotton-tipped swabs as described previously?[14],?[15]. As the 12 restrooms were nearly identical in design, we were able to swab the same area at each location between restrooms. In order to characterize tap water communities as a potential source of bacteria, 1 L of faucet water from six of the restrooms (each building having the same water source for each restroom sampled) was collected and filtered through 0.2 ?m bottle top filters (Nalgene, Rochester, NY, USA)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	22132229	NA	NA	NA	Gilberto Flores	CCME	GAZ:United States of America	NA	NA	0	NA	NA	NA	NA	n	NA	Noah Fierer	Flores_restroom_surface_biogeography	pyrosequencing	NA	NA	NA	restroom surface metagenome	3	ENVO:city	Titanium	NA	NA	NA	NA	NA	flores_restroom_surface_biogeography	restroom surface biogeography	256318	CCME	NA	NA	NA	1335	"Genomic DNA was extracted from the swabs and filters using the MO BIO PowerSoil DNA isolation kit following the manufacturer's protocol with the modifications of Fierer?et al.?[14]. A portion of the 16 S rRNA gene spanning the V1_V2 regions was amplified using the primer set (27 F/338R), PCR mixture conditions and thermal cycling conditions described in Fierer?et al.?[15]. PCR amplicons of triplicate reactions for each sample were pooled at approximately equal amounts and pyrosequenced at 454 Life Sciences (Branford, CT, USA) on their GS Junior system. A total of 337,333 high-quality partial 16 S rRNA gene sequences were obtained from 101 of the 120 surface samples collected, averaging approximately 3,340 sequences per sample (ranging from 513_6,771) (Table S1) in 4 GS Junior runs, with the best run containing 116,004 high-quality reads. An additional 16,416 sequences (ranging from 2161_5084 per sample) were generated for five of the six water samples collected for source tracking analysis. Each sample was amplified with a unique barcode to enable multiplexing in the GS Junior runs. The barcoded sequencing reads can be separated by data analysis software providing high confidence in assigning sequencing read to each sample. Sequence data generated as part of this study is available upon request by contacting the corresponding author."	NA	NA	y	NA	NA	ENVO:surface	16S rRNA	Ekeley	NA	V2	CCME	NA	Stall out	NA	NA	NA	40.01499	NA	NA	Door lock area out of stall
PTAM4.489517	GCACTCGTTAGA	YATGCTGCCTCCCGTAGGAGT	jesse.stombaugh@colorado.edu	NA	restroom surface	Door_lock_area_out_of_stall	NA	NA	NA	n	12/13/10	restroom surface metagenome	NA	GFL0IUQ	NA	mimarks-survey	NA	NA	NA	NA	0	NA	NA	1624.1	10-Nov	Male	0	NA	NA	NA	NA	NA	NA	NA	NA	NA	NA	"1,swab"	A	NA	Microbial biogeography of public restroom surfaces	NA	NA	-105.27055	y	NA	CCME	NA	NA	NA	"We spend the majority of our lives indoors where we are constantly exposed to bacteria residing on surfaces. However, the diversity of these surface-associated communities is largely unknown. We explored the biogeographical patterns exhibited by bacteria across ten surfaces within each of twelve public restrooms. Using high-throughput barcoded pyrosequencing of the 16S rRNA gene, we identified 19 bacterial phyla across all surfaces. Most sequences belonged to four phyla: Actinobacteria, Bacteriodetes, Firmicutes and Proteobacteria. The communities clustered into three general categories: those found on surfaces associated with toilets, those on the restroom floor, and those found on surfaces routinely touched with hands. On toilet surfaces, gut-associated taxa were more prevalent, suggesting fecal contamination of these surfaces. Floor surfaces were the most diverse of all communities and contained several taxa commonly found in soils. Skin-associated bacteria, especially the Propionibacteriaceae, dominated surfaces routinely touched with our hands. Certain taxa were more common in female than in male restrooms as vagina-associated Lactobacillaceae were widely distributed in female restrooms, likely from urine contamination. Use of the SourceTracker algorithm confirmed many of our taxonomic observations as human skin was the primary source of bacteria on restroom surfaces. Overall, these results demonstrate that restroom surfaces host relatively diverse microbial communities dominated by human-associated bacteria with clear linkages between communities on or in different body sites and those communities found on restroom surfaces. More generally, this work is relevant to the public health field as we show that human associated microbes are commonly found on restroom surfaces suggesting that bacterial pathogens could readily be transmitted between individuals by the touching of surfaces. Furthermore, we demonstrate that we can use high-throughput analyses of bacterial communities to determine sources of bacteria on indoor surfaces, an approach which could be used to track pathogen transmission and test the efficacy of hygiene practices."	ENVO:building	TCAG	Microbial biogeography of public restroom surfaces	NA	0	"Ten surfaces (door handles into and out of the restroom, handles into and out of a restroom stall, faucet handles, soap dispenser, toilet seat, toilet flush handle, floor around the toilet and floor around the sink) in six male and six female restrooms evenly distributed across two buildings on the University of Colorado at Boulder campus were sampled on a single day in November 2010. Surfaces where sampled using sterile, cotton-tipped swabs as described previously?[14],?[15]. As the 12 restrooms were nearly identical in design, we were able to swab the same area at each location between restrooms. In order to characterize tap water communities as a potential source of bacteria, 1 L of faucet water from six of the restrooms (each building having the same water source for each restroom sampled) was collected and filtered through 0.2 ?m bottle top filters (Nalgene, Rochester, NY, USA)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	22132229	NA	NA	NA	Gilberto Flores	CCME	GAZ:United States of America	NA	NA	0	NA	NA	NA	NA	n	NA	Noah Fierer	Flores_restroom_surface_biogeography	pyrosequencing	NA	NA	NA	restroom surface metagenome	1	ENVO:city	Titanium	NA	NA	NA	NA	NA	flores_restroom_surface_biogeography	restroom surface biogeography	256318	CCME	NA	NA	NA	1335	"Genomic DNA was extracted from the swabs and filters using the MO BIO PowerSoil DNA isolation kit following the manufacturer's protocol with the modifications of Fierer?et al.?[14]. A portion of the 16 S rRNA gene spanning the V1_V2 regions was amplified using the primer set (27 F/338R), PCR mixture conditions and thermal cycling conditions described in Fierer?et al.?[15]. PCR amplicons of triplicate reactions for each sample were pooled at approximately equal amounts and pyrosequenced at 454 Life Sciences (Branford, CT, USA) on their GS Junior system. A total of 337,333 high-quality partial 16 S rRNA gene sequences were obtained from 101 of the 120 surface samples collected, averaging approximately 3,340 sequences per sample (ranging from 513_6,771) (Table S1) in 4 GS Junior runs, with the best run containing 116,004 high-quality reads. An additional 16,416 sequences (ranging from 2161_5084 per sample) were generated for five of the six water samples collected for source tracking analysis. Each sample was amplified with a unique barcode to enable multiplexing in the GS Junior runs. The barcoded sequencing reads can be separated by data analysis software providing high confidence in assigning sequencing read to each sample. Sequence data generated as part of this study is available upon request by contacting the corresponding author."	NA	NA	y	NA	NA	ENVO:surface	16S rRNA	Porter	NA	V2	CCME	NA	Stall out	NA	NA	NA	40.01499	NA	NA	Door lock area out of stall
PTBM4.489490	GCGTACAACTGT	YATGCTGCCTCCCGTAGGAGT	jesse.stombaugh@colorado.edu	NA	restroom surface	Door_lock_area_out_of_stall	NA	NA	NA	n	12/13/10	restroom surface metagenome	NA	GFL0IUQ	NA	mimarks-survey	NA	NA	NA	NA	0	NA	NA	1624.1	10-Nov	Male	0	NA	NA	NA	NA	NA	NA	NA	NA	NA	NA	"1,swab"	B	NA	Microbial biogeography of public restroom surfaces	NA	NA	-105.27055	y	NA	CCME	NA	NA	NA	"We spend the majority of our lives indoors where we are constantly exposed to bacteria residing on surfaces. However, the diversity of these surface-associated communities is largely unknown. We explored the biogeographical patterns exhibited by bacteria across ten surfaces within each of twelve public restrooms. Using high-throughput barcoded pyrosequencing of the 16S rRNA gene, we identified 19 bacterial phyla across all surfaces. Most sequences belonged to four phyla: Actinobacteria, Bacteriodetes, Firmicutes and Proteobacteria. The communities clustered into three general categories: those found on surfaces associated with toilets, those on the restroom floor, and those found on surfaces routinely touched with hands. On toilet surfaces, gut-associated taxa were more prevalent, suggesting fecal contamination of these surfaces. Floor surfaces were the most diverse of all communities and contained several taxa commonly found in soils. Skin-associated bacteria, especially the Propionibacteriaceae, dominated surfaces routinely touched with our hands. Certain taxa were more common in female than in male restrooms as vagina-associated Lactobacillaceae were widely distributed in female restrooms, likely from urine contamination. Use of the SourceTracker algorithm confirmed many of our taxonomic observations as human skin was the primary source of bacteria on restroom surfaces. Overall, these results demonstrate that restroom surfaces host relatively diverse microbial communities dominated by human-associated bacteria with clear linkages between communities on or in different body sites and those communities found on restroom surfaces. More generally, this work is relevant to the public health field as we show that human associated microbes are commonly found on restroom surfaces suggesting that bacterial pathogens could readily be transmitted between individuals by the touching of surfaces. Furthermore, we demonstrate that we can use high-throughput analyses of bacterial communities to determine sources of bacteria on indoor surfaces, an approach which could be used to track pathogen transmission and test the efficacy of hygiene practices."	ENVO:building	TCAG	Microbial biogeography of public restroom surfaces	NA	0	"Ten surfaces (door handles into and out of the restroom, handles into and out of a restroom stall, faucet handles, soap dispenser, toilet seat, toilet flush handle, floor around the toilet and floor around the sink) in six male and six female restrooms evenly distributed across two buildings on the University of Colorado at Boulder campus were sampled on a single day in November 2010. Surfaces where sampled using sterile, cotton-tipped swabs as described previously?[14],?[15]. As the 12 restrooms were nearly identical in design, we were able to swab the same area at each location between restrooms. In order to characterize tap water communities as a potential source of bacteria, 1 L of faucet water from six of the restrooms (each building having the same water source for each restroom sampled) was collected and filtered through 0.2 ?m bottle top filters (Nalgene, Rochester, NY, USA)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	22132229	NA	NA	NA	Gilberto Flores	CCME	GAZ:United States of America	NA	NA	0	NA	NA	NA	NA	n	NA	Noah Fierer	Flores_restroom_surface_biogeography	pyrosequencing	NA	NA	NA	restroom surface metagenome	2	ENVO:city	Titanium	NA	NA	NA	NA	NA	flores_restroom_surface_biogeography	restroom surface biogeography	256318	CCME	NA	NA	NA	1335	"Genomic DNA was extracted from the swabs and filters using the MO BIO PowerSoil DNA isolation kit following the manufacturer's protocol with the modifications of Fierer?et al.?[14]. A portion of the 16 S rRNA gene spanning the V1_V2 regions was amplified using the primer set (27 F/338R), PCR mixture conditions and thermal cycling conditions described in Fierer?et al.?[15]. PCR amplicons of triplicate reactions for each sample were pooled at approximately equal amounts and pyrosequenced at 454 Life Sciences (Branford, CT, USA) on their GS Junior system. A total of 337,333 high-quality partial 16 S rRNA gene sequences were obtained from 101 of the 120 surface samples collected, averaging approximately 3,340 sequences per sample (ranging from 513_6,771) (Table S1) in 4 GS Junior runs, with the best run containing 116,004 high-quality reads. An additional 16,416 sequences (ranging from 2161_5084 per sample) were generated for five of the six water samples collected for source tracking analysis. Each sample was amplified with a unique barcode to enable multiplexing in the GS Junior runs. The barcoded sequencing reads can be separated by data analysis software providing high confidence in assigning sequencing read to each sample. Sequence data generated as part of this study is available upon request by contacting the corresponding author."	NA	NA	y	NA	NA	ENVO:surface	16S rRNA	Porter	NA	V2	CCME	NA	Stall out	NA	NA	NA	40.01499	NA	NA	Door lock area out of stall
PTCF4.489538	ATCACGTAGCGG	YATGCTGCCTCCCGTAGGAGT	jesse.stombaugh@colorado.edu	NA	restroom surface	Door_lock_area_out_of_stall	NA	NA	NA	n	12/13/10	restroom surface metagenome	NA	GFL0IUQ	NA	mimarks-survey	NA	NA	NA	NA	0	NA	NA	1624.1	10-Nov	Female	0	NA	NA	NA	NA	NA	NA	NA	NA	NA	NA	"1,swab"	C	NA	Microbial biogeography of public restroom surfaces	NA	NA	-105.27055	y	NA	CCME	NA	NA	NA	"We spend the majority of our lives indoors where we are constantly exposed to bacteria residing on surfaces. However, the diversity of these surface-associated communities is largely unknown. We explored the biogeographical patterns exhibited by bacteria across ten surfaces within each of twelve public restrooms. Using high-throughput barcoded pyrosequencing of the 16S rRNA gene, we identified 19 bacterial phyla across all surfaces. Most sequences belonged to four phyla: Actinobacteria, Bacteriodetes, Firmicutes and Proteobacteria. The communities clustered into three general categories: those found on surfaces associated with toilets, those on the restroom floor, and those found on surfaces routinely touched with hands. On toilet surfaces, gut-associated taxa were more prevalent, suggesting fecal contamination of these surfaces. Floor surfaces were the most diverse of all communities and contained several taxa commonly found in soils. Skin-associated bacteria, especially the Propionibacteriaceae, dominated surfaces routinely touched with our hands. Certain taxa were more common in female than in male restrooms as vagina-associated Lactobacillaceae were widely distributed in female restrooms, likely from urine contamination. Use of the SourceTracker algorithm confirmed many of our taxonomic observations as human skin was the primary source of bacteria on restroom surfaces. Overall, these results demonstrate that restroom surfaces host relatively diverse microbial communities dominated by human-associated bacteria with clear linkages between communities on or in different body sites and those communities found on restroom surfaces. More generally, this work is relevant to the public health field as we show that human associated microbes are commonly found on restroom surfaces suggesting that bacterial pathogens could readily be transmitted between individuals by the touching of surfaces. Furthermore, we demonstrate that we can use high-throughput analyses of bacterial communities to determine sources of bacteria on indoor surfaces, an approach which could be used to track pathogen transmission and test the efficacy of hygiene practices."	ENVO:building	TCAG	Microbial biogeography of public restroom surfaces	NA	0	"Ten surfaces (door handles into and out of the restroom, handles into and out of a restroom stall, faucet handles, soap dispenser, toilet seat, toilet flush handle, floor around the toilet and floor around the sink) in six male and six female restrooms evenly distributed across two buildings on the University of Colorado at Boulder campus were sampled on a single day in November 2010. Surfaces where sampled using sterile, cotton-tipped swabs as described previously?[14],?[15]. As the 12 restrooms were nearly identical in design, we were able to swab the same area at each location between restrooms. In order to characterize tap water communities as a potential source of bacteria, 1 L of faucet water from six of the restrooms (each building having the same water source for each restroom sampled) was collected and filtered through 0.2 ?m bottle top filters (Nalgene, Rochester, NY, USA)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	22132229	NA	NA	NA	Gilberto Flores	CCME	GAZ:United States of America	NA	NA	0	NA	NA	NA	NA	n	NA	Noah Fierer	Flores_restroom_surface_biogeography	pyrosequencing	NA	NA	NA	restroom surface metagenome	3	ENVO:city	Titanium	NA	NA	NA	NA	NA	flores_restroom_surface_biogeography	restroom surface biogeography	256318	CCME	NA	NA	NA	1335	"Genomic DNA was extracted from the swabs and filters using the MO BIO PowerSoil DNA isolation kit following the manufacturer's protocol with the modifications of Fierer?et al.?[14]. A portion of the 16 S rRNA gene spanning the V1_V2 regions was amplified using the primer set (27 F/338R), PCR mixture conditions and thermal cycling conditions described in Fierer?et al.?[15]. PCR amplicons of triplicate reactions for each sample were pooled at approximately equal amounts and pyrosequenced at 454 Life Sciences (Branford, CT, USA) on their GS Junior system. A total of 337,333 high-quality partial 16 S rRNA gene sequences were obtained from 101 of the 120 surface samples collected, averaging approximately 3,340 sequences per sample (ranging from 513_6,771) (Table S1) in 4 GS Junior runs, with the best run containing 116,004 high-quality reads. An additional 16,416 sequences (ranging from 2161_5084 per sample) were generated for five of the six water samples collected for source tracking analysis. Each sample was amplified with a unique barcode to enable multiplexing in the GS Junior runs. The barcoded sequencing reads can be separated by data analysis software providing high confidence in assigning sequencing read to each sample. Sequence data generated as part of this study is available upon request by contacting the corresponding author."	NA	NA	y	NA	NA	ENVO:surface	16S rRNA	Porter	NA	V2	CCME	NA	Stall out	NA	NA	NA	40.01499	NA	NA	Door lock area out of stall
PTCM4.489484	ATAATCTCGTCG	YATGCTGCCTCCCGTAGGAGT	jesse.stombaugh@colorado.edu	NA	restroom surface	Door_lock_area_out_of_stall	NA	NA	NA	n	12/13/10	restroom surface metagenome	NA	GFL0IUQ	NA	mimarks-survey	NA	NA	NA	NA	0	NA	NA	1624.1	10-Nov	Male	0	NA	NA	NA	NA	NA	NA	NA	NA	NA	NA	"1,swab"	C	NA	Microbial biogeography of public restroom surfaces	NA	NA	-105.27055	y	NA	CCME	NA	NA	NA	"We spend the majority of our lives indoors where we are constantly exposed to bacteria residing on surfaces. However, the diversity of these surface-associated communities is largely unknown. We explored the biogeographical patterns exhibited by bacteria across ten surfaces within each of twelve public restrooms. Using high-throughput barcoded pyrosequencing of the 16S rRNA gene, we identified 19 bacterial phyla across all surfaces. Most sequences belonged to four phyla: Actinobacteria, Bacteriodetes, Firmicutes and Proteobacteria. The communities clustered into three general categories: those found on surfaces associated with toilets, those on the restroom floor, and those found on surfaces routinely touched with hands. On toilet surfaces, gut-associated taxa were more prevalent, suggesting fecal contamination of these surfaces. Floor surfaces were the most diverse of all communities and contained several taxa commonly found in soils. Skin-associated bacteria, especially the Propionibacteriaceae, dominated surfaces routinely touched with our hands. Certain taxa were more common in female than in male restrooms as vagina-associated Lactobacillaceae were widely distributed in female restrooms, likely from urine contamination. Use of the SourceTracker algorithm confirmed many of our taxonomic observations as human skin was the primary source of bacteria on restroom surfaces. Overall, these results demonstrate that restroom surfaces host relatively diverse microbial communities dominated by human-associated bacteria with clear linkages between communities on or in different body sites and those communities found on restroom surfaces. More generally, this work is relevant to the public health field as we show that human associated microbes are commonly found on restroom surfaces suggesting that bacterial pathogens could readily be transmitted between individuals by the touching of surfaces. Furthermore, we demonstrate that we can use high-throughput analyses of bacterial communities to determine sources of bacteria on indoor surfaces, an approach which could be used to track pathogen transmission and test the efficacy of hygiene practices."	ENVO:building	TCAG	Microbial biogeography of public restroom surfaces	NA	0	"Ten surfaces (door handles into and out of the restroom, handles into and out of a restroom stall, faucet handles, soap dispenser, toilet seat, toilet flush handle, floor around the toilet and floor around the sink) in six male and six female restrooms evenly distributed across two buildings on the University of Colorado at Boulder campus were sampled on a single day in November 2010. Surfaces where sampled using sterile, cotton-tipped swabs as described previously?[14],?[15]. As the 12 restrooms were nearly identical in design, we were able to swab the same area at each location between restrooms. In order to characterize tap water communities as a potential source of bacteria, 1 L of faucet water from six of the restrooms (each building having the same water source for each restroom sampled) was collected and filtered through 0.2 ?m bottle top filters (Nalgene, Rochester, NY, USA)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	22132229	NA	NA	NA	Gilberto Flores	CCME	GAZ:United States of America	NA	NA	0	NA	NA	NA	NA	n	NA	Noah Fierer	Flores_restroom_surface_biogeography	pyrosequencing	NA	NA	NA	restroom surface metagenome	3	ENVO:city	Titanium	NA	NA	NA	NA	NA	flores_restroom_surface_biogeography	restroom surface biogeography	256318	CCME	NA	NA	NA	1335	"Genomic DNA was extracted from the swabs and filters using the MO BIO PowerSoil DNA isolation kit following the manufacturer's protocol with the modifications of Fierer?et al.?[14]. A portion of the 16 S rRNA gene spanning the V1_V2 regions was amplified using the primer set (27 F/338R), PCR mixture conditions and thermal cycling conditions described in Fierer?et al.?[15]. PCR amplicons of triplicate reactions for each sample were pooled at approximately equal amounts and pyrosequenced at 454 Life Sciences (Branford, CT, USA) on their GS Junior system. A total of 337,333 high-quality partial 16 S rRNA gene sequences were obtained from 101 of the 120 surface samples collected, averaging approximately 3,340 sequences per sample (ranging from 513_6,771) (Table S1) in 4 GS Junior runs, with the best run containing 116,004 high-quality reads. An additional 16,416 sequences (ranging from 2161_5084 per sample) were generated for five of the six water samples collected for source tracking analysis. Each sample was amplified with a unique barcode to enable multiplexing in the GS Junior runs. The barcoded sequencing reads can be separated by data analysis software providing high confidence in assigning sequencing read to each sample. Sequence data generated as part of this study is available upon request by contacting the corresponding author."	NA	NA	y	NA	NA	ENVO:surface	16S rRNA	Porter	NA	V2	CCME	NA	Stall out	NA	NA	NA	40.01499	NA	NA	Door lock area out of stall
EKAF8.489558	GACGAGTCAGTC	YATGCTGCCTCCCGTAGGAGT	jesse.stombaugh@colorado.edu	NA	restroom surface	Faucet_handles	NA	NA	NA	n	12/13/10	restroom surface metagenome	NA	GFL0IUQ	NA	mimarks-survey	NA	NA	NA	NA	0	NA	NA	1624.1	10-Nov	Female	0	NA	NA	NA	NA	NA	NA	NA	NA	NA	NA	"1,swab"	A	NA	Microbial biogeography of public restroom surfaces	NA	NA	-105.27055	y	NA	CCME	NA	NA	NA	"We spend the majority of our lives indoors where we are constantly exposed to bacteria residing on surfaces. However, the diversity of these surface-associated communities is largely unknown. We explored the biogeographical patterns exhibited by bacteria across ten surfaces within each of twelve public restrooms. Using high-throughput barcoded pyrosequencing of the 16S rRNA gene, we identified 19 bacterial phyla across all surfaces. Most sequences belonged to four phyla: Actinobacteria, Bacteriodetes, Firmicutes and Proteobacteria. The communities clustered into three general categories: those found on surfaces associated with toilets, those on the restroom floor, and those found on surfaces routinely touched with hands. On toilet surfaces, gut-associated taxa were more prevalent, suggesting fecal contamination of these surfaces. Floor surfaces were the most diverse of all communities and contained several taxa commonly found in soils. Skin-associated bacteria, especially the Propionibacteriaceae, dominated surfaces routinely touched with our hands. Certain taxa were more common in female than in male restrooms as vagina-associated Lactobacillaceae were widely distributed in female restrooms, likely from urine contamination. Use of the SourceTracker algorithm confirmed many of our taxonomic observations as human skin was the primary source of bacteria on restroom surfaces. Overall, these results demonstrate that restroom surfaces host relatively diverse microbial communities dominated by human-associated bacteria with clear linkages between communities on or in different body sites and those communities found on restroom surfaces. More generally, this work is relevant to the public health field as we show that human associated microbes are commonly found on restroom surfaces suggesting that bacterial pathogens could readily be transmitted between individuals by the touching of surfaces. Furthermore, we demonstrate that we can use high-throughput analyses of bacterial communities to determine sources of bacteria on indoor surfaces, an approach which could be used to track pathogen transmission and test the efficacy of hygiene practices."	ENVO:building	TCAG	Microbial biogeography of public restroom surfaces	NA	0	"Ten surfaces (door handles into and out of the restroom, handles into and out of a restroom stall, faucet handles, soap dispenser, toilet seat, toilet flush handle, floor around the toilet and floor around the sink) in six male and six female restrooms evenly distributed across two buildings on the University of Colorado at Boulder campus were sampled on a single day in November 2010. Surfaces where sampled using sterile, cotton-tipped swabs as described previously?[14],?[15]. As the 12 restrooms were nearly identical in design, we were able to swab the same area at each location between restrooms. In order to characterize tap water communities as a potential source of bacteria, 1 L of faucet water from six of the restrooms (each building having the same water source for each restroom sampled) was collected and filtered through 0.2 ?m bottle top filters (Nalgene, Rochester, NY, USA)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	22132229	NA	NA	NA	Gilberto Flores	CCME	GAZ:United States of America	NA	NA	0	NA	NA	NA	NA	n	NA	Noah Fierer	Flores_restroom_surface_biogeography	pyrosequencing	NA	NA	NA	restroom surface metagenome	1	ENVO:city	Titanium	NA	NA	NA	NA	NA	flores_restroom_surface_biogeography	restroom surface biogeography	256318	CCME	NA	NA	NA	1335	"Genomic DNA was extracted from the swabs and filters using the MO BIO PowerSoil DNA isolation kit following the manufacturer's protocol with the modifications of Fierer?et al.?[14]. A portion of the 16 S rRNA gene spanning the V1_V2 regions was amplified using the primer set (27 F/338R), PCR mixture conditions and thermal cycling conditions described in Fierer?et al.?[15]. PCR amplicons of triplicate reactions for each sample were pooled at approximately equal amounts and pyrosequenced at 454 Life Sciences (Branford, CT, USA) on their GS Junior system. A total of 337,333 high-quality partial 16 S rRNA gene sequences were obtained from 101 of the 120 surface samples collected, averaging approximately 3,340 sequences per sample (ranging from 513_6,771) (Table S1) in 4 GS Junior runs, with the best run containing 116,004 high-quality reads. An additional 16,416 sequences (ranging from 2161_5084 per sample) were generated for five of the six water samples collected for source tracking analysis. Each sample was amplified with a unique barcode to enable multiplexing in the GS Junior runs. The barcoded sequencing reads can be separated by data analysis software providing high confidence in assigning sequencing read to each sample. Sequence data generated as part of this study is available upon request by contacting the corresponding author."	NA	NA	y	NA	NA	ENVO:surface	16S rRNA	Ekeley	NA	V2	CCME	NA	Faucet handles	NA	NA	NA	40.01499	NA	NA	Faucet handles
EKAM8.489462	GAAGCTACTGTC	YATGCTGCCTCCCGTAGGAGT	jesse.stombaugh@colorado.edu	NA	restroom surface	Faucet_handles	NA	NA	NA	n	12/13/10	restroom surface metagenome	NA	GFL0IUQ	NA	mimarks-survey	NA	NA	NA	NA	0	NA	NA	1624.1	10-Nov	Male	0	NA	NA	NA	NA	NA	NA	NA	NA	NA	NA	"1,swab"	A	NA	Microbial biogeography of public restroom surfaces	NA	NA	-105.27055	y	NA	CCME	NA	NA	NA	"We spend the majority of our lives indoors where we are constantly exposed to bacteria residing on surfaces. However, the diversity of these surface-associated communities is largely unknown. We explored the biogeographical patterns exhibited by bacteria across ten surfaces within each of twelve public restrooms. Using high-throughput barcoded pyrosequencing of the 16S rRNA gene, we identified 19 bacterial phyla across all surfaces. Most sequences belonged to four phyla: Actinobacteria, Bacteriodetes, Firmicutes and Proteobacteria. The communities clustered into three general categories: those found on surfaces associated with toilets, those on the restroom floor, and those found on surfaces routinely touched with hands. On toilet surfaces, gut-associated taxa were more prevalent, suggesting fecal contamination of these surfaces. Floor surfaces were the most diverse of all communities and contained several taxa commonly found in soils. Skin-associated bacteria, especially the Propionibacteriaceae, dominated surfaces routinely touched with our hands. Certain taxa were more common in female than in male restrooms as vagina-associated Lactobacillaceae were widely distributed in female restrooms, likely from urine contamination. Use of the SourceTracker algorithm confirmed many of our taxonomic observations as human skin was the primary source of bacteria on restroom surfaces. Overall, these results demonstrate that restroom surfaces host relatively diverse microbial communities dominated by human-associated bacteria with clear linkages between communities on or in different body sites and those communities found on restroom surfaces. More generally, this work is relevant to the public health field as we show that human associated microbes are commonly found on restroom surfaces suggesting that bacterial pathogens could readily be transmitted between individuals by the touching of surfaces. Furthermore, we demonstrate that we can use high-throughput analyses of bacterial communities to determine sources of bacteria on indoor surfaces, an approach which could be used to track pathogen transmission and test the efficacy of hygiene practices."	ENVO:building	TCAG	Microbial biogeography of public restroom surfaces	NA	0	"Ten surfaces (door handles into and out of the restroom, handles into and out of a restroom stall, faucet handles, soap dispenser, toilet seat, toilet flush handle, floor around the toilet and floor around the sink) in six male and six female restrooms evenly distributed across two buildings on the University of Colorado at Boulder campus were sampled on a single day in November 2010. Surfaces where sampled using sterile, cotton-tipped swabs as described previously?[14],?[15]. As the 12 restrooms were nearly identical in design, we were able to swab the same area at each location between restrooms. In order to characterize tap water communities as a potential source of bacteria, 1 L of faucet water from six of the restrooms (each building having the same water source for each restroom sampled) was collected and filtered through 0.2 ?m bottle top filters (Nalgene, Rochester, NY, USA)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	22132229	NA	NA	NA	Gilberto Flores	CCME	GAZ:United States of America	NA	NA	0	NA	NA	NA	NA	n	NA	Noah Fierer	Flores_restroom_surface_biogeography	pyrosequencing	NA	NA	NA	restroom surface metagenome	1	ENVO:city	Titanium	NA	NA	NA	NA	NA	flores_restroom_surface_biogeography	restroom surface biogeography	256318	CCME	NA	NA	NA	1335	"Genomic DNA was extracted from the swabs and filters using the MO BIO PowerSoil DNA isolation kit following the manufacturer's protocol with the modifications of Fierer?et al.?[14]. A portion of the 16 S rRNA gene spanning the V1_V2 regions was amplified using the primer set (27 F/338R), PCR mixture conditions and thermal cycling conditions described in Fierer?et al.?[15]. PCR amplicons of triplicate reactions for each sample were pooled at approximately equal amounts and pyrosequenced at 454 Life Sciences (Branford, CT, USA) on their GS Junior system. A total of 337,333 high-quality partial 16 S rRNA gene sequences were obtained from 101 of the 120 surface samples collected, averaging approximately 3,340 sequences per sample (ranging from 513_6,771) (Table S1) in 4 GS Junior runs, with the best run containing 116,004 high-quality reads. An additional 16,416 sequences (ranging from 2161_5084 per sample) were generated for five of the six water samples collected for source tracking analysis. Each sample was amplified with a unique barcode to enable multiplexing in the GS Junior runs. The barcoded sequencing reads can be separated by data analysis software providing high confidence in assigning sequencing read to each sample. Sequence data generated as part of this study is available upon request by contacting the corresponding author."	NA	NA	y	NA	NA	ENVO:surface	16S rRNA	Ekeley	NA	V2	CCME	NA	Faucet handles	NA	NA	NA	40.01499	NA	NA	Faucet handles
EKBF8.489493	GAGCTGGCTGAT	YATGCTGCCTCCCGTAGGAGT	jesse.stombaugh@colorado.edu	NA	restroom surface	Faucet_handles	NA	NA	NA	n	12/13/10	restroom surface metagenome	NA	GFL0IUQ	NA	mimarks-survey	NA	NA	NA	NA	0	NA	NA	1624.1	10-Nov	Female	0	NA	NA	NA	NA	NA	NA	NA	NA	NA	NA	"1,swab"	B	NA	Microbial biogeography of public restroom surfaces	NA	NA	-105.27055	y	NA	CCME	NA	NA	NA	"We spend the majority of our lives indoors where we are constantly exposed to bacteria residing on surfaces. However, the diversity of these surface-associated communities is largely unknown. We explored the biogeographical patterns exhibited by bacteria across ten surfaces within each of twelve public restrooms. Using high-throughput barcoded pyrosequencing of the 16S rRNA gene, we identified 19 bacterial phyla across all surfaces. Most sequences belonged to four phyla: Actinobacteria, Bacteriodetes, Firmicutes and Proteobacteria. The communities clustered into three general categories: those found on surfaces associated with toilets, those on the restroom floor, and those found on surfaces routinely touched with hands. On toilet surfaces, gut-associated taxa were more prevalent, suggesting fecal contamination of these surfaces. Floor surfaces were the most diverse of all communities and contained several taxa commonly found in soils. Skin-associated bacteria, especially the Propionibacteriaceae, dominated surfaces routinely touched with our hands. Certain taxa were more common in female than in male restrooms as vagina-associated Lactobacillaceae were widely distributed in female restrooms, likely from urine contamination. Use of the SourceTracker algorithm confirmed many of our taxonomic observations as human skin was the primary source of bacteria on restroom surfaces. Overall, these results demonstrate that restroom surfaces host relatively diverse microbial communities dominated by human-associated bacteria with clear linkages between communities on or in different body sites and those communities found on restroom surfaces. More generally, this work is relevant to the public health field as we show that human associated microbes are commonly found on restroom surfaces suggesting that bacterial pathogens could readily be transmitted between individuals by the touching of surfaces. Furthermore, we demonstrate that we can use high-throughput analyses of bacterial communities to determine sources of bacteria on indoor surfaces, an approach which could be used to track pathogen transmission and test the efficacy of hygiene practices."	ENVO:building	TCAG	Microbial biogeography of public restroom surfaces	NA	0	"Ten surfaces (door handles into and out of the restroom, handles into and out of a restroom stall, faucet handles, soap dispenser, toilet seat, toilet flush handle, floor around the toilet and floor around the sink) in six male and six female restrooms evenly distributed across two buildings on the University of Colorado at Boulder campus were sampled on a single day in November 2010. Surfaces where sampled using sterile, cotton-tipped swabs as described previously?[14],?[15]. As the 12 restrooms were nearly identical in design, we were able to swab the same area at each location between restrooms. In order to characterize tap water communities as a potential source of bacteria, 1 L of faucet water from six of the restrooms (each building having the same water source for each restroom sampled) was collected and filtered through 0.2 ?m bottle top filters (Nalgene, Rochester, NY, USA)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	22132229	NA	NA	NA	Gilberto Flores	CCME	GAZ:United States of America	NA	NA	0	NA	NA	NA	NA	n	NA	Noah Fierer	Flores_restroom_surface_biogeography	pyrosequencing	NA	NA	NA	restroom surface metagenome	2	ENVO:city	Titanium	NA	NA	NA	NA	NA	flores_restroom_surface_biogeography	restroom surface biogeography	256318	CCME	NA	NA	NA	1335	"Genomic DNA was extracted from the swabs and filters using the MO BIO PowerSoil DNA isolation kit following the manufacturer's protocol with the modifications of Fierer?et al.?[14]. A portion of the 16 S rRNA gene spanning the V1_V2 regions was amplified using the primer set (27 F/338R), PCR mixture conditions and thermal cycling conditions described in Fierer?et al.?[15]. PCR amplicons of triplicate reactions for each sample were pooled at approximately equal amounts and pyrosequenced at 454 Life Sciences (Branford, CT, USA) on their GS Junior system. A total of 337,333 high-quality partial 16 S rRNA gene sequences were obtained from 101 of the 120 surface samples collected, averaging approximately 3,340 sequences per sample (ranging from 513_6,771) (Table S1) in 4 GS Junior runs, with the best run containing 116,004 high-quality reads. An additional 16,416 sequences (ranging from 2161_5084 per sample) were generated for five of the six water samples collected for source tracking analysis. Each sample was amplified with a unique barcode to enable multiplexing in the GS Junior runs. The barcoded sequencing reads can be separated by data analysis software providing high confidence in assigning sequencing read to each sample. Sequence data generated as part of this study is available upon request by contacting the corresponding author."	NA	NA	y	NA	NA	ENVO:surface	16S rRNA	Ekeley	NA	V2	CCME	NA	Faucet handles	NA	NA	NA	40.01499	NA	NA	Faucet handles
EKBM8.489473	GACTGCATCTTA	YATGCTGCCTCCCGTAGGAGT	jesse.stombaugh@colorado.edu	NA	restroom surface	Faucet_handles	NA	NA	NA	n	12/13/10	restroom surface metagenome	NA	GFL0IUQ	NA	mimarks-survey	NA	NA	NA	NA	0	NA	NA	1624.1	10-Nov	Male	0	NA	NA	NA	NA	NA	NA	NA	NA	NA	NA	"1,swab"	B	NA	Microbial biogeography of public restroom surfaces	NA	NA	-105.27055	y	NA	CCME	NA	NA	NA	"We spend the majority of our lives indoors where we are constantly exposed to bacteria residing on surfaces. However, the diversity of these surface-associated communities is largely unknown. We explored the biogeographical patterns exhibited by bacteria across ten surfaces within each of twelve public restrooms. Using high-throughput barcoded pyrosequencing of the 16S rRNA gene, we identified 19 bacterial phyla across all surfaces. Most sequences belonged to four phyla: Actinobacteria, Bacteriodetes, Firmicutes and Proteobacteria. The communities clustered into three general categories: those found on surfaces associated with toilets, those on the restroom floor, and those found on surfaces routinely touched with hands. On toilet surfaces, gut-associated taxa were more prevalent, suggesting fecal contamination of these surfaces. Floor surfaces were the most diverse of all communities and contained several taxa commonly found in soils. Skin-associated bacteria, especially the Propionibacteriaceae, dominated surfaces routinely touched with our hands. Certain taxa were more common in female than in male restrooms as vagina-associated Lactobacillaceae were widely distributed in female restrooms, likely from urine contamination. Use of the SourceTracker algorithm confirmed many of our taxonomic observations as human skin was the primary source of bacteria on restroom surfaces. Overall, these results demonstrate that restroom surfaces host relatively diverse microbial communities dominated by human-associated bacteria with clear linkages between communities on or in different body sites and those communities found on restroom surfaces. More generally, this work is relevant to the public health field as we show that human associated microbes are commonly found on restroom surfaces suggesting that bacterial pathogens could readily be transmitted between individuals by the touching of surfaces. Furthermore, we demonstrate that we can use high-throughput analyses of bacterial communities to determine sources of bacteria on indoor surfaces, an approach which could be used to track pathogen transmission and test the efficacy of hygiene practices."	ENVO:building	TCAG	Microbial biogeography of public restroom surfaces	NA	0	"Ten surfaces (door handles into and out of the restroom, handles into and out of a restroom stall, faucet handles, soap dispenser, toilet seat, toilet flush handle, floor around the toilet and floor around the sink) in six male and six female restrooms evenly distributed across two buildings on the University of Colorado at Boulder campus were sampled on a single day in November 2010. Surfaces where sampled using sterile, cotton-tipped swabs as described previously?[14],?[15]. As the 12 restrooms were nearly identical in design, we were able to swab the same area at each location between restrooms. In order to characterize tap water communities as a potential source of bacteria, 1 L of faucet water from six of the restrooms (each building having the same water source for each restroom sampled) was collected and filtered through 0.2 ?m bottle top filters (Nalgene, Rochester, NY, USA)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	22132229	NA	NA	NA	Gilberto Flores	CCME	GAZ:United States of America	NA	NA	0	NA	NA	NA	NA	n	NA	Noah Fierer	Flores_restroom_surface_biogeography	pyrosequencing	NA	NA	NA	restroom surface metagenome	2	ENVO:city	Titanium	NA	NA	NA	NA	NA	flores_restroom_surface_biogeography	restroom surface biogeography	256318	CCME	NA	NA	NA	1335	"Genomic DNA was extracted from the swabs and filters using the MO BIO PowerSoil DNA isolation kit following the manufacturer's protocol with the modifications of Fierer?et al.?[14]. A portion of the 16 S rRNA gene spanning the V1_V2 regions was amplified using the primer set (27 F/338R), PCR mixture conditions and thermal cycling conditions described in Fierer?et al.?[15]. PCR amplicons of triplicate reactions for each sample were pooled at approximately equal amounts and pyrosequenced at 454 Life Sciences (Branford, CT, USA) on their GS Junior system. A total of 337,333 high-quality partial 16 S rRNA gene sequences were obtained from 101 of the 120 surface samples collected, averaging approximately 3,340 sequences per sample (ranging from 513_6,771) (Table S1) in 4 GS Junior runs, with the best run containing 116,004 high-quality reads. An additional 16,416 sequences (ranging from 2161_5084 per sample) were generated for five of the six water samples collected for source tracking analysis. Each sample was amplified with a unique barcode to enable multiplexing in the GS Junior runs. The barcoded sequencing reads can be separated by data analysis software providing high confidence in assigning sequencing read to each sample. Sequence data generated as part of this study is available upon request by contacting the corresponding author."	NA	NA	y	NA	NA	ENVO:surface	16S rRNA	Ekeley	NA	V2	CCME	NA	Faucet handles	NA	NA	NA	40.01499	NA	NA	Faucet handles
EKCF8.489457	GATGTCGTGTCA	YATGCTGCCTCCCGTAGGAGT	jesse.stombaugh@colorado.edu	NA	restroom surface	Faucet_handles	NA	NA	NA	n	12/13/10	restroom surface metagenome	NA	GFL0IUQ	NA	mimarks-survey	NA	NA	NA	NA	0	NA	NA	1624.1	10-Nov	Female	0	NA	NA	NA	NA	NA	NA	NA	NA	NA	NA	"1,swab"	C	NA	Microbial biogeography of public restroom surfaces	NA	NA	-105.27055	y	NA	CCME	NA	NA	NA	"We spend the majority of our lives indoors where we are constantly exposed to bacteria residing on surfaces. However, the diversity of these surface-associated communities is largely unknown. We explored the biogeographical patterns exhibited by bacteria across ten surfaces within each of twelve public restrooms. Using high-throughput barcoded pyrosequencing of the 16S rRNA gene, we identified 19 bacterial phyla across all surfaces. Most sequences belonged to four phyla: Actinobacteria, Bacteriodetes, Firmicutes and Proteobacteria. The communities clustered into three general categories: those found on surfaces associated with toilets, those on the restroom floor, and those found on surfaces routinely touched with hands. On toilet surfaces, gut-associated taxa were more prevalent, suggesting fecal contamination of these surfaces. Floor surfaces were the most diverse of all communities and contained several taxa commonly found in soils. Skin-associated bacteria, especially the Propionibacteriaceae, dominated surfaces routinely touched with our hands. Certain taxa were more common in female than in male restrooms as vagina-associated Lactobacillaceae were widely distributed in female restrooms, likely from urine contamination. Use of the SourceTracker algorithm confirmed many of our taxonomic observations as human skin was the primary source of bacteria on restroom surfaces. Overall, these results demonstrate that restroom surfaces host relatively diverse microbial communities dominated by human-associated bacteria with clear linkages between communities on or in different body sites and those communities found on restroom surfaces. More generally, this work is relevant to the public health field as we show that human associated microbes are commonly found on restroom surfaces suggesting that bacterial pathogens could readily be transmitted between individuals by the touching of surfaces. Furthermore, we demonstrate that we can use high-throughput analyses of bacterial communities to determine sources of bacteria on indoor surfaces, an approach which could be used to track pathogen transmission and test the efficacy of hygiene practices."	ENVO:building	TCAG	Microbial biogeography of public restroom surfaces	NA	0	"Ten surfaces (door handles into and out of the restroom, handles into and out of a restroom stall, faucet handles, soap dispenser, toilet seat, toilet flush handle, floor around the toilet and floor around the sink) in six male and six female restrooms evenly distributed across two buildings on the University of Colorado at Boulder campus were sampled on a single day in November 2010. Surfaces where sampled using sterile, cotton-tipped swabs as described previously?[14],?[15]. As the 12 restrooms were nearly identical in design, we were able to swab the same area at each location between restrooms. In order to characterize tap water communities as a potential source of bacteria, 1 L of faucet water from six of the restrooms (each building having the same water source for each restroom sampled) was collected and filtered through 0.2 ?m bottle top filters (Nalgene, Rochester, NY, USA)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	22132229	NA	NA	NA	Gilberto Flores	CCME	GAZ:United States of America	NA	NA	0	NA	NA	NA	NA	n	NA	Noah Fierer	Flores_restroom_surface_biogeography	pyrosequencing	NA	NA	NA	restroom surface metagenome	3	ENVO:city	Titanium	NA	NA	NA	NA	NA	flores_restroom_surface_biogeography	restroom surface biogeography	256318	CCME	NA	NA	NA	1335	"Genomic DNA was extracted from the swabs and filters using the MO BIO PowerSoil DNA isolation kit following the manufacturer's protocol with the modifications of Fierer?et al.?[14]. A portion of the 16 S rRNA gene spanning the V1_V2 regions was amplified using the primer set (27 F/338R), PCR mixture conditions and thermal cycling conditions described in Fierer?et al.?[15]. PCR amplicons of triplicate reactions for each sample were pooled at approximately equal amounts and pyrosequenced at 454 Life Sciences (Branford, CT, USA) on their GS Junior system. A total of 337,333 high-quality partial 16 S rRNA gene sequences were obtained from 101 of the 120 surface samples collected, averaging approximately 3,340 sequences per sample (ranging from 513_6,771) (Table S1) in 4 GS Junior runs, with the best run containing 116,004 high-quality reads. An additional 16,416 sequences (ranging from 2161_5084 per sample) were generated for five of the six water samples collected for source tracking analysis. Each sample was amplified with a unique barcode to enable multiplexing in the GS Junior runs. The barcoded sequencing reads can be separated by data analysis software providing high confidence in assigning sequencing read to each sample. Sequence data generated as part of this study is available upon request by contacting the corresponding author."	NA	NA	y	NA	NA	ENVO:surface	16S rRNA	Ekeley	NA	V2	CCME	NA	Faucet handles	NA	NA	NA	40.01499	NA	NA	Faucet handles
EKCM8.489573	GATATGCGGCTG	YATGCTGCCTCCCGTAGGAGT	jesse.stombaugh@colorado.edu	NA	restroom surface	Faucet_handles	NA	NA	NA	n	12/13/10	restroom surface metagenome	NA	GFL0IUQ	NA	mimarks-survey	NA	NA	NA	NA	0	NA	NA	1624.1	10-Nov	Male	0	NA	NA	NA	NA	NA	NA	NA	NA	NA	NA	"1,swab"	C	NA	Microbial biogeography of public restroom surfaces	NA	NA	-105.27055	y	NA	CCME	NA	NA	NA	"We spend the majority of our lives indoors where we are constantly exposed to bacteria residing on surfaces. However, the diversity of these surface-associated communities is largely unknown. We explored the biogeographical patterns exhibited by bacteria across ten surfaces within each of twelve public restrooms. Using high-throughput barcoded pyrosequencing of the 16S rRNA gene, we identified 19 bacterial phyla across all surfaces. Most sequences belonged to four phyla: Actinobacteria, Bacteriodetes, Firmicutes and Proteobacteria. The communities clustered into three general categories: those found on surfaces associated with toilets, those on the restroom floor, and those found on surfaces routinely touched with hands. On toilet surfaces, gut-associated taxa were more prevalent, suggesting fecal contamination of these surfaces. Floor surfaces were the most diverse of all communities and contained several taxa commonly found in soils. Skin-associated bacteria, especially the Propionibacteriaceae, dominated surfaces routinely touched with our hands. Certain taxa were more common in female than in male restrooms as vagina-associated Lactobacillaceae were widely distributed in female restrooms, likely from urine contamination. Use of the SourceTracker algorithm confirmed many of our taxonomic observations as human skin was the primary source of bacteria on restroom surfaces. Overall, these results demonstrate that restroom surfaces host relatively diverse microbial communities dominated by human-associated bacteria with clear linkages between communities on or in different body sites and those communities found on restroom surfaces. More generally, this work is relevant to the public health field as we show that human associated microbes are commonly found on restroom surfaces suggesting that bacterial pathogens could readily be transmitted between individuals by the touching of surfaces. Furthermore, we demonstrate that we can use high-throughput analyses of bacterial communities to determine sources of bacteria on indoor surfaces, an approach which could be used to track pathogen transmission and test the efficacy of hygiene practices."	ENVO:building	TCAG	Microbial biogeography of public restroom surfaces	NA	0	"Ten surfaces (door handles into and out of the restroom, handles into and out of a restroom stall, faucet handles, soap dispenser, toilet seat, toilet flush handle, floor around the toilet and floor around the sink) in six male and six female restrooms evenly distributed across two buildings on the University of Colorado at Boulder campus were sampled on a single day in November 2010. Surfaces where sampled using sterile, cotton-tipped swabs as described previously?[14],?[15]. As the 12 restrooms were nearly identical in design, we were able to swab the same area at each location between restrooms. In order to characterize tap water communities as a potential source of bacteria, 1 L of faucet water from six of the restrooms (each building having the same water source for each restroom sampled) was collected and filtered through 0.2 ?m bottle top filters (Nalgene, Rochester, NY, USA)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	22132229	NA	NA	NA	Gilberto Flores	CCME	GAZ:United States of America	NA	NA	0	NA	NA	NA	NA	n	NA	Noah Fierer	Flores_restroom_surface_biogeography	pyrosequencing	NA	NA	NA	restroom surface metagenome	3	ENVO:city	Titanium	NA	NA	NA	NA	NA	flores_restroom_surface_biogeography	restroom surface biogeography	256318	CCME	NA	NA	NA	1335	"Genomic DNA was extracted from the swabs and filters using the MO BIO PowerSoil DNA isolation kit following the manufacturer's protocol with the modifications of Fierer?et al.?[14]. A portion of the 16 S rRNA gene spanning the V1_V2 regions was amplified using the primer set (27 F/338R), PCR mixture conditions and thermal cycling conditions described in Fierer?et al.?[15]. PCR amplicons of triplicate reactions for each sample were pooled at approximately equal amounts and pyrosequenced at 454 Life Sciences (Branford, CT, USA) on their GS Junior system. A total of 337,333 high-quality partial 16 S rRNA gene sequences were obtained from 101 of the 120 surface samples collected, averaging approximately 3,340 sequences per sample (ranging from 513_6,771) (Table S1) in 4 GS Junior runs, with the best run containing 116,004 high-quality reads. An additional 16,416 sequences (ranging from 2161_5084 per sample) were generated for five of the six water samples collected for source tracking analysis. Each sample was amplified with a unique barcode to enable multiplexing in the GS Junior runs. The barcoded sequencing reads can be separated by data analysis software providing high confidence in assigning sequencing read to each sample. Sequence data generated as part of this study is available upon request by contacting the corresponding author."	NA	NA	y	NA	NA	ENVO:surface	16S rRNA	Ekeley	NA	V2	CCME	NA	Faucet handles	NA	NA	NA	40.01499	NA	NA	Faucet handles
PTAF8.489480	GCCAGAGTCGTA	YATGCTGCCTCCCGTAGGAGT	jesse.stombaugh@colorado.edu	NA	restroom surface	Faucet_handles	NA	NA	NA	n	12/13/10	restroom surface metagenome	NA	GFL0IUQ	NA	mimarks-survey	NA	NA	NA	NA	0	NA	NA	1624.1	10-Nov	Female	0	NA	NA	NA	NA	NA	NA	NA	NA	NA	NA	"1,swab"	A	NA	Microbial biogeography of public restroom surfaces	NA	NA	-105.27055	y	NA	CCME	NA	NA	NA	"We spend the majority of our lives indoors where we are constantly exposed to bacteria residing on surfaces. However, the diversity of these surface-associated communities is largely unknown. We explored the biogeographical patterns exhibited by bacteria across ten surfaces within each of twelve public restrooms. Using high-throughput barcoded pyrosequencing of the 16S rRNA gene, we identified 19 bacterial phyla across all surfaces. Most sequences belonged to four phyla: Actinobacteria, Bacteriodetes, Firmicutes and Proteobacteria. The communities clustered into three general categories: those found on surfaces associated with toilets, those on the restroom floor, and those found on surfaces routinely touched with hands. On toilet surfaces, gut-associated taxa were more prevalent, suggesting fecal contamination of these surfaces. Floor surfaces were the most diverse of all communities and contained several taxa commonly found in soils. Skin-associated bacteria, especially the Propionibacteriaceae, dominated surfaces routinely touched with our hands. Certain taxa were more common in female than in male restrooms as vagina-associated Lactobacillaceae were widely distributed in female restrooms, likely from urine contamination. Use of the SourceTracker algorithm confirmed many of our taxonomic observations as human skin was the primary source of bacteria on restroom surfaces. Overall, these results demonstrate that restroom surfaces host relatively diverse microbial communities dominated by human-associated bacteria with clear linkages between communities on or in different body sites and those communities found on restroom surfaces. More generally, this work is relevant to the public health field as we show that human associated microbes are commonly found on restroom surfaces suggesting that bacterial pathogens could readily be transmitted between individuals by the touching of surfaces. Furthermore, we demonstrate that we can use high-throughput analyses of bacterial communities to determine sources of bacteria on indoor surfaces, an approach which could be used to track pathogen transmission and test the efficacy of hygiene practices."	ENVO:building	TCAG	Microbial biogeography of public restroom surfaces	NA	0	"Ten surfaces (door handles into and out of the restroom, handles into and out of a restroom stall, faucet handles, soap dispenser, toilet seat, toilet flush handle, floor around the toilet and floor around the sink) in six male and six female restrooms evenly distributed across two buildings on the University of Colorado at Boulder campus were sampled on a single day in November 2010. Surfaces where sampled using sterile, cotton-tipped swabs as described previously?[14],?[15]. As the 12 restrooms were nearly identical in design, we were able to swab the same area at each location between restrooms. In order to characterize tap water communities as a potential source of bacteria, 1 L of faucet water from six of the restrooms (each building having the same water source for each restroom sampled) was collected and filtered through 0.2 ?m bottle top filters (Nalgene, Rochester, NY, USA)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	22132229	NA	NA	NA	Gilberto Flores	CCME	GAZ:United States of America	NA	NA	0	NA	NA	NA	NA	n	NA	Noah Fierer	Flores_restroom_surface_biogeography	pyrosequencing	NA	NA	NA	restroom surface metagenome	1	ENVO:city	Titanium	NA	NA	NA	NA	NA	flores_restroom_surface_biogeography	restroom surface biogeography	256318	CCME	NA	NA	NA	1335	"Genomic DNA was extracted from the swabs and filters using the MO BIO PowerSoil DNA isolation kit following the manufacturer's protocol with the modifications of Fierer?et al.?[14]. A portion of the 16 S rRNA gene spanning the V1_V2 regions was amplified using the primer set (27 F/338R), PCR mixture conditions and thermal cycling conditions described in Fierer?et al.?[15]. PCR amplicons of triplicate reactions for each sample were pooled at approximately equal amounts and pyrosequenced at 454 Life Sciences (Branford, CT, USA) on their GS Junior system. A total of 337,333 high-quality partial 16 S rRNA gene sequences were obtained from 101 of the 120 surface samples collected, averaging approximately 3,340 sequences per sample (ranging from 513_6,771) (Table S1) in 4 GS Junior runs, with the best run containing 116,004 high-quality reads. An additional 16,416 sequences (ranging from 2161_5084 per sample) were generated for five of the six water samples collected for source tracking analysis. Each sample was amplified with a unique barcode to enable multiplexing in the GS Junior runs. The barcoded sequencing reads can be separated by data analysis software providing high confidence in assigning sequencing read to each sample. Sequence data generated as part of this study is available upon request by contacting the corresponding author."	NA	NA	y	NA	NA	ENVO:surface	16S rRNA	Porter	NA	V2	CCME	NA	Faucet handles	NA	NA	NA	40.01499	NA	NA	Faucet handles
PTAM8.489453	GCAGGATAGATA	YATGCTGCCTCCCGTAGGAGT	jesse.stombaugh@colorado.edu	NA	restroom surface	Faucet_handles	NA	NA	NA	n	12/13/10	restroom surface metagenome	NA	GFL0IUQ	NA	mimarks-survey	NA	NA	NA	NA	0	NA	NA	1624.1	10-Nov	Male	0	NA	NA	NA	NA	NA	NA	NA	NA	NA	NA	"1,swab"	A	NA	Microbial biogeography of public restroom surfaces	NA	NA	-105.27055	y	NA	CCME	NA	NA	NA	"We spend the majority of our lives indoors where we are constantly exposed to bacteria residing on surfaces. However, the diversity of these surface-associated communities is largely unknown. We explored the biogeographical patterns exhibited by bacteria across ten surfaces within each of twelve public restrooms. Using high-throughput barcoded pyrosequencing of the 16S rRNA gene, we identified 19 bacterial phyla across all surfaces. Most sequences belonged to four phyla: Actinobacteria, Bacteriodetes, Firmicutes and Proteobacteria. The communities clustered into three general categories: those found on surfaces associated with toilets, those on the restroom floor, and those found on surfaces routinely touched with hands. On toilet surfaces, gut-associated taxa were more prevalent, suggesting fecal contamination of these surfaces. Floor surfaces were the most diverse of all communities and contained several taxa commonly found in soils. Skin-associated bacteria, especially the Propionibacteriaceae, dominated surfaces routinely touched with our hands. Certain taxa were more common in female than in male restrooms as vagina-associated Lactobacillaceae were widely distributed in female restrooms, likely from urine contamination. Use of the SourceTracker algorithm confirmed many of our taxonomic observations as human skin was the primary source of bacteria on restroom surfaces. Overall, these results demonstrate that restroom surfaces host relatively diverse microbial communities dominated by human-associated bacteria with clear linkages between communities on or in different body sites and those communities found on restroom surfaces. More generally, this work is relevant to the public health field as we show that human associated microbes are commonly found on restroom surfaces suggesting that bacterial pathogens could readily be transmitted between individuals by the touching of surfaces. Furthermore, we demonstrate that we can use high-throughput analyses of bacterial communities to determine sources of bacteria on indoor surfaces, an approach which could be used to track pathogen transmission and test the efficacy of hygiene practices."	ENVO:building	TCAG	Microbial biogeography of public restroom surfaces	NA	0	"Ten surfaces (door handles into and out of the restroom, handles into and out of a restroom stall, faucet handles, soap dispenser, toilet seat, toilet flush handle, floor around the toilet and floor around the sink) in six male and six female restrooms evenly distributed across two buildings on the University of Colorado at Boulder campus were sampled on a single day in November 2010. Surfaces where sampled using sterile, cotton-tipped swabs as described previously?[14],?[15]. As the 12 restrooms were nearly identical in design, we were able to swab the same area at each location between restrooms. In order to characterize tap water communities as a potential source of bacteria, 1 L of faucet water from six of the restrooms (each building having the same water source for each restroom sampled) was collected and filtered through 0.2 ?m bottle top filters (Nalgene, Rochester, NY, USA)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	22132229	NA	NA	NA	Gilberto Flores	CCME	GAZ:United States of America	NA	NA	0	NA	NA	NA	NA	n	NA	Noah Fierer	Flores_restroom_surface_biogeography	pyrosequencing	NA	NA	NA	restroom surface metagenome	1	ENVO:city	Titanium	NA	NA	NA	NA	NA	flores_restroom_surface_biogeography	restroom surface biogeography	256318	CCME	NA	NA	NA	1335	"Genomic DNA was extracted from the swabs and filters using the MO BIO PowerSoil DNA isolation kit following the manufacturer's protocol with the modifications of Fierer?et al.?[14]. A portion of the 16 S rRNA gene spanning the V1_V2 regions was amplified using the primer set (27 F/338R), PCR mixture conditions and thermal cycling conditions described in Fierer?et al.?[15]. PCR amplicons of triplicate reactions for each sample were pooled at approximately equal amounts and pyrosequenced at 454 Life Sciences (Branford, CT, USA) on their GS Junior system. A total of 337,333 high-quality partial 16 S rRNA gene sequences were obtained from 101 of the 120 surface samples collected, averaging approximately 3,340 sequences per sample (ranging from 513_6,771) (Table S1) in 4 GS Junior runs, with the best run containing 116,004 high-quality reads. An additional 16,416 sequences (ranging from 2161_5084 per sample) were generated for five of the six water samples collected for source tracking analysis. Each sample was amplified with a unique barcode to enable multiplexing in the GS Junior runs. The barcoded sequencing reads can be separated by data analysis software providing high confidence in assigning sequencing read to each sample. Sequence data generated as part of this study is available upon request by contacting the corresponding author."	NA	NA	y	NA	NA	ENVO:surface	16S rRNA	Porter	NA	V2	CCME	NA	Faucet handles	NA	NA	NA	40.01499	NA	NA	Faucet handles
PTBM8.489497	GCTAGATGCCAG	YATGCTGCCTCCCGTAGGAGT	jesse.stombaugh@colorado.edu	NA	restroom surface	Faucet_handles	NA	NA	NA	n	12/13/10	restroom surface metagenome	NA	GFL0IUQ	NA	mimarks-survey	NA	NA	NA	NA	0	NA	NA	1624.1	10-Nov	Male	0	NA	NA	NA	NA	NA	NA	NA	NA	NA	NA	"1,swab"	B	NA	Microbial biogeography of public restroom surfaces	NA	NA	-105.27055	y	NA	CCME	NA	NA	NA	"We spend the majority of our lives indoors where we are constantly exposed to bacteria residing on surfaces. However, the diversity of these surface-associated communities is largely unknown. We explored the biogeographical patterns exhibited by bacteria across ten surfaces within each of twelve public restrooms. Using high-throughput barcoded pyrosequencing of the 16S rRNA gene, we identified 19 bacterial phyla across all surfaces. Most sequences belonged to four phyla: Actinobacteria, Bacteriodetes, Firmicutes and Proteobacteria. The communities clustered into three general categories: those found on surfaces associated with toilets, those on the restroom floor, and those found on surfaces routinely touched with hands. On toilet surfaces, gut-associated taxa were more prevalent, suggesting fecal contamination of these surfaces. Floor surfaces were the most diverse of all communities and contained several taxa commonly found in soils. Skin-associated bacteria, especially the Propionibacteriaceae, dominated surfaces routinely touched with our hands. Certain taxa were more common in female than in male restrooms as vagina-associated Lactobacillaceae were widely distributed in female restrooms, likely from urine contamination. Use of the SourceTracker algorithm confirmed many of our taxonomic observations as human skin was the primary source of bacteria on restroom surfaces. Overall, these results demonstrate that restroom surfaces host relatively diverse microbial communities dominated by human-associated bacteria with clear linkages between communities on or in different body sites and those communities found on restroom surfaces. More generally, this work is relevant to the public health field as we show that human associated microbes are commonly found on restroom surfaces suggesting that bacterial pathogens could readily be transmitted between individuals by the touching of surfaces. Furthermore, we demonstrate that we can use high-throughput analyses of bacterial communities to determine sources of bacteria on indoor surfaces, an approach which could be used to track pathogen transmission and test the efficacy of hygiene practices."	ENVO:building	TCAG	Microbial biogeography of public restroom surfaces	NA	0	"Ten surfaces (door handles into and out of the restroom, handles into and out of a restroom stall, faucet handles, soap dispenser, toilet seat, toilet flush handle, floor around the toilet and floor around the sink) in six male and six female restrooms evenly distributed across two buildings on the University of Colorado at Boulder campus were sampled on a single day in November 2010. Surfaces where sampled using sterile, cotton-tipped swabs as described previously?[14],?[15]. As the 12 restrooms were nearly identical in design, we were able to swab the same area at each location between restrooms. In order to characterize tap water communities as a potential source of bacteria, 1 L of faucet water from six of the restrooms (each building having the same water source for each restroom sampled) was collected and filtered through 0.2 ?m bottle top filters (Nalgene, Rochester, NY, USA)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	22132229	NA	NA	NA	Gilberto Flores	CCME	GAZ:United States of America	NA	NA	0	NA	NA	NA	NA	n	NA	Noah Fierer	Flores_restroom_surface_biogeography	pyrosequencing	NA	NA	NA	restroom surface metagenome	2	ENVO:city	Titanium	NA	NA	NA	NA	NA	flores_restroom_surface_biogeography	restroom surface biogeography	256318	CCME	NA	NA	NA	1335	"Genomic DNA was extracted from the swabs and filters using the MO BIO PowerSoil DNA isolation kit following the manufacturer's protocol with the modifications of Fierer?et al.?[14]. A portion of the 16 S rRNA gene spanning the V1_V2 regions was amplified using the primer set (27 F/338R), PCR mixture conditions and thermal cycling conditions described in Fierer?et al.?[15]. PCR amplicons of triplicate reactions for each sample were pooled at approximately equal amounts and pyrosequenced at 454 Life Sciences (Branford, CT, USA) on their GS Junior system. A total of 337,333 high-quality partial 16 S rRNA gene sequences were obtained from 101 of the 120 surface samples collected, averaging approximately 3,340 sequences per sample (ranging from 513_6,771) (Table S1) in 4 GS Junior runs, with the best run containing 116,004 high-quality reads. An additional 16,416 sequences (ranging from 2161_5084 per sample) were generated for five of the six water samples collected for source tracking analysis. Each sample was amplified with a unique barcode to enable multiplexing in the GS Junior runs. The barcoded sequencing reads can be separated by data analysis software providing high confidence in assigning sequencing read to each sample. Sequence data generated as part of this study is available upon request by contacting the corresponding author."	NA	NA	y	NA	NA	ENVO:surface	16S rRNA	Porter	NA	V2	CCME	NA	Faucet handles	NA	NA	NA	40.01499	NA	NA	Faucet handles
PTCF8.489486	ATCCTCAGTAGT	YATGCTGCCTCCCGTAGGAGT	jesse.stombaugh@colorado.edu	NA	restroom surface	Faucet_handles	NA	NA	NA	n	12/13/10	restroom surface metagenome	NA	GFL0IUQ	NA	mimarks-survey	NA	NA	NA	NA	0	NA	NA	1624.1	10-Nov	Female	0	NA	NA	NA	NA	NA	NA	NA	NA	NA	NA	"1,swab"	C	NA	Microbial biogeography of public restroom surfaces	NA	NA	-105.27055	y	NA	CCME	NA	NA	NA	"We spend the majority of our lives indoors where we are constantly exposed to bacteria residing on surfaces. However, the diversity of these surface-associated communities is largely unknown. We explored the biogeographical patterns exhibited by bacteria across ten surfaces within each of twelve public restrooms. Using high-throughput barcoded pyrosequencing of the 16S rRNA gene, we identified 19 bacterial phyla across all surfaces. Most sequences belonged to four phyla: Actinobacteria, Bacteriodetes, Firmicutes and Proteobacteria. The communities clustered into three general categories: those found on surfaces associated with toilets, those on the restroom floor, and those found on surfaces routinely touched with hands. On toilet surfaces, gut-associated taxa were more prevalent, suggesting fecal contamination of these surfaces. Floor surfaces were the most diverse of all communities and contained several taxa commonly found in soils. Skin-associated bacteria, especially the Propionibacteriaceae, dominated surfaces routinely touched with our hands. Certain taxa were more common in female than in male restrooms as vagina-associated Lactobacillaceae were widely distributed in female restrooms, likely from urine contamination. Use of the SourceTracker algorithm confirmed many of our taxonomic observations as human skin was the primary source of bacteria on restroom surfaces. Overall, these results demonstrate that restroom surfaces host relatively diverse microbial communities dominated by human-associated bacteria with clear linkages between communities on or in different body sites and those communities found on restroom surfaces. More generally, this work is relevant to the public health field as we show that human associated microbes are commonly found on restroom surfaces suggesting that bacterial pathogens could readily be transmitted between individuals by the touching of surfaces. Furthermore, we demonstrate that we can use high-throughput analyses of bacterial communities to determine sources of bacteria on indoor surfaces, an approach which could be used to track pathogen transmission and test the efficacy of hygiene practices."	ENVO:building	TCAG	Microbial biogeography of public restroom surfaces	NA	0	"Ten surfaces (door handles into and out of the restroom, handles into and out of a restroom stall, faucet handles, soap dispenser, toilet seat, toilet flush handle, floor around the toilet and floor around the sink) in six male and six female restrooms evenly distributed across two buildings on the University of Colorado at Boulder campus were sampled on a single day in November 2010. Surfaces where sampled using sterile, cotton-tipped swabs as described previously?[14],?[15]. As the 12 restrooms were nearly identical in design, we were able to swab the same area at each location between restrooms. In order to characterize tap water communities as a potential source of bacteria, 1 L of faucet water from six of the restrooms (each building having the same water source for each restroom sampled) was collected and filtered through 0.2 ?m bottle top filters (Nalgene, Rochester, NY, USA)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	22132229	NA	NA	NA	Gilberto Flores	CCME	GAZ:United States of America	NA	NA	0	NA	NA	NA	NA	n	NA	Noah Fierer	Flores_restroom_surface_biogeography	pyrosequencing	NA	NA	NA	restroom surface metagenome	3	ENVO:city	Titanium	NA	NA	NA	NA	NA	flores_restroom_surface_biogeography	restroom surface biogeography	256318	CCME	NA	NA	NA	1335	"Genomic DNA was extracted from the swabs and filters using the MO BIO PowerSoil DNA isolation kit following the manufacturer's protocol with the modifications of Fierer?et al.?[14]. A portion of the 16 S rRNA gene spanning the V1_V2 regions was amplified using the primer set (27 F/338R), PCR mixture conditions and thermal cycling conditions described in Fierer?et al.?[15]. PCR amplicons of triplicate reactions for each sample were pooled at approximately equal amounts and pyrosequenced at 454 Life Sciences (Branford, CT, USA) on their GS Junior system. A total of 337,333 high-quality partial 16 S rRNA gene sequences were obtained from 101 of the 120 surface samples collected, averaging approximately 3,340 sequences per sample (ranging from 513_6,771) (Table S1) in 4 GS Junior runs, with the best run containing 116,004 high-quality reads. An additional 16,416 sequences (ranging from 2161_5084 per sample) were generated for five of the six water samples collected for source tracking analysis. Each sample was amplified with a unique barcode to enable multiplexing in the GS Junior runs. The barcoded sequencing reads can be separated by data analysis software providing high confidence in assigning sequencing read to each sample. Sequence data generated as part of this study is available upon request by contacting the corresponding author."	NA	NA	y	NA	NA	ENVO:surface	16S rRNA	Porter	NA	V2	CCME	NA	Faucet handles	NA	NA	NA	40.01499	NA	NA	Faucet handles
PTCM8.489465	ATACTATTGCGC	YATGCTGCCTCCCGTAGGAGT	jesse.stombaugh@colorado.edu	NA	restroom surface	Faucet_handles	NA	NA	NA	n	12/13/10	restroom surface metagenome	NA	GFL0IUQ	NA	mimarks-survey	NA	NA	NA	NA	0	NA	NA	1624.1	10-Nov	Male	0	NA	NA	NA	NA	NA	NA	NA	NA	NA	NA	"1,swab"	C	NA	Microbial biogeography of public restroom surfaces	NA	NA	-105.27055	y	NA	CCME	NA	NA	NA	"We spend the majority of our lives indoors where we are constantly exposed to bacteria residing on surfaces. However, the diversity of these surface-associated communities is largely unknown. We explored the biogeographical patterns exhibited by bacteria across ten surfaces within each of twelve public restrooms. Using high-throughput barcoded pyrosequencing of the 16S rRNA gene, we identified 19 bacterial phyla across all surfaces. Most sequences belonged to four phyla: Actinobacteria, Bacteriodetes, Firmicutes and Proteobacteria. The communities clustered into three general categories: those found on surfaces associated with toilets, those on the restroom floor, and those found on surfaces routinely touched with hands. On toilet surfaces, gut-associated taxa were more prevalent, suggesting fecal contamination of these surfaces. Floor surfaces were the most diverse of all communities and contained several taxa commonly found in soils. Skin-associated bacteria, especially the Propionibacteriaceae, dominated surfaces routinely touched with our hands. Certain taxa were more common in female than in male restrooms as vagina-associated Lactobacillaceae were widely distributed in female restrooms, likely from urine contamination. Use of the SourceTracker algorithm confirmed many of our taxonomic observations as human skin was the primary source of bacteria on restroom surfaces. Overall, these results demonstrate that restroom surfaces host relatively diverse microbial communities dominated by human-associated bacteria with clear linkages between communities on or in different body sites and those communities found on restroom surfaces. More generally, this work is relevant to the public health field as we show that human associated microbes are commonly found on restroom surfaces suggesting that bacterial pathogens could readily be transmitted between individuals by the touching of surfaces. Furthermore, we demonstrate that we can use high-throughput analyses of bacterial communities to determine sources of bacteria on indoor surfaces, an approach which could be used to track pathogen transmission and test the efficacy of hygiene practices."	ENVO:building	TCAG	Microbial biogeography of public restroom surfaces	NA	0	"Ten surfaces (door handles into and out of the restroom, handles into and out of a restroom stall, faucet handles, soap dispenser, toilet seat, toilet flush handle, floor around the toilet and floor around the sink) in six male and six female restrooms evenly distributed across two buildings on the University of Colorado at Boulder campus were sampled on a single day in November 2010. Surfaces where sampled using sterile, cotton-tipped swabs as described previously?[14],?[15]. As the 12 restrooms were nearly identical in design, we were able to swab the same area at each location between restrooms. In order to characterize tap water communities as a potential source of bacteria, 1 L of faucet water from six of the restrooms (each building having the same water source for each restroom sampled) was collected and filtered through 0.2 ?m bottle top filters (Nalgene, Rochester, NY, USA)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	22132229	NA	NA	NA	Gilberto Flores	CCME	GAZ:United States of America	NA	NA	0	NA	NA	NA	NA	n	NA	Noah Fierer	Flores_restroom_surface_biogeography	pyrosequencing	NA	NA	NA	restroom surface metagenome	3	ENVO:city	Titanium	NA	NA	NA	NA	NA	flores_restroom_surface_biogeography	restroom surface biogeography	256318	CCME	NA	NA	NA	1335	"Genomic DNA was extracted from the swabs and filters using the MO BIO PowerSoil DNA isolation kit following the manufacturer's protocol with the modifications of Fierer?et al.?[14]. A portion of the 16 S rRNA gene spanning the V1_V2 regions was amplified using the primer set (27 F/338R), PCR mixture conditions and thermal cycling conditions described in Fierer?et al.?[15]. PCR amplicons of triplicate reactions for each sample were pooled at approximately equal amounts and pyrosequenced at 454 Life Sciences (Branford, CT, USA) on their GS Junior system. A total of 337,333 high-quality partial 16 S rRNA gene sequences were obtained from 101 of the 120 surface samples collected, averaging approximately 3,340 sequences per sample (ranging from 513_6,771) (Table S1) in 4 GS Junior runs, with the best run containing 116,004 high-quality reads. An additional 16,416 sequences (ranging from 2161_5084 per sample) were generated for five of the six water samples collected for source tracking analysis. Each sample was amplified with a unique barcode to enable multiplexing in the GS Junior runs. The barcoded sequencing reads can be separated by data analysis software providing high confidence in assigning sequencing read to each sample. Sequence data generated as part of this study is available upon request by contacting the corresponding author."	NA	NA	y	NA	NA	ENVO:surface	16S rRNA	Porter	NA	V2	CCME	NA	Faucet handles	NA	NA	NA	40.01499	NA	NA	Faucet handles
EKAF10.489503	GACGCAGTAGCT	YATGCTGCCTCCCGTAGGAGT	jesse.stombaugh@colorado.edu	NA	restroom surface	Floor_between_sink_and_door	NA	NA	NA	n	12/13/10	restroom surface metagenome	NA	GFL0IUQ	NA	mimarks-survey	NA	NA	NA	NA	0	NA	NA	1624.1	10-Nov	Female	0	NA	NA	NA	NA	NA	NA	NA	NA	NA	NA	"1,swab"	A	NA	Microbial biogeography of public restroom surfaces	NA	NA	-105.27055	y	NA	CCME	NA	NA	NA	"We spend the majority of our lives indoors where we are constantly exposed to bacteria residing on surfaces. However, the diversity of these surface-associated communities is largely unknown. We explored the biogeographical patterns exhibited by bacteria across ten surfaces within each of twelve public restrooms. Using high-throughput barcoded pyrosequencing of the 16S rRNA gene, we identified 19 bacterial phyla across all surfaces. Most sequences belonged to four phyla: Actinobacteria, Bacteriodetes, Firmicutes and Proteobacteria. The communities clustered into three general categories: those found on surfaces associated with toilets, those on the restroom floor, and those found on surfaces routinely touched with hands. On toilet surfaces, gut-associated taxa were more prevalent, suggesting fecal contamination of these surfaces. Floor surfaces were the most diverse of all communities and contained several taxa commonly found in soils. Skin-associated bacteria, especially the Propionibacteriaceae, dominated surfaces routinely touched with our hands. Certain taxa were more common in female than in male restrooms as vagina-associated Lactobacillaceae were widely distributed in female restrooms, likely from urine contamination. Use of the SourceTracker algorithm confirmed many of our taxonomic observations as human skin was the primary source of bacteria on restroom surfaces. Overall, these results demonstrate that restroom surfaces host relatively diverse microbial communities dominated by human-associated bacteria with clear linkages between communities on or in different body sites and those communities found on restroom surfaces. More generally, this work is relevant to the public health field as we show that human associated microbes are commonly found on restroom surfaces suggesting that bacterial pathogens could readily be transmitted between individuals by the touching of surfaces. Furthermore, we demonstrate that we can use high-throughput analyses of bacterial communities to determine sources of bacteria on indoor surfaces, an approach which could be used to track pathogen transmission and test the efficacy of hygiene practices."	ENVO:building	TCAG	Microbial biogeography of public restroom surfaces	NA	0	"Ten surfaces (door handles into and out of the restroom, handles into and out of a restroom stall, faucet handles, soap dispenser, toilet seat, toilet flush handle, floor around the toilet and floor around the sink) in six male and six female restrooms evenly distributed across two buildings on the University of Colorado at Boulder campus were sampled on a single day in November 2010. Surfaces where sampled using sterile, cotton-tipped swabs as described previously?[14],?[15]. As the 12 restrooms were nearly identical in design, we were able to swab the same area at each location between restrooms. In order to characterize tap water communities as a potential source of bacteria, 1 L of faucet water from six of the restrooms (each building having the same water source for each restroom sampled) was collected and filtered through 0.2 ?m bottle top filters (Nalgene, Rochester, NY, USA)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	22132229	NA	NA	NA	Gilberto Flores	CCME	GAZ:United States of America	NA	NA	0	NA	NA	NA	NA	n	NA	Noah Fierer	Flores_restroom_surface_biogeography	pyrosequencing	NA	NA	NA	restroom surface metagenome	1	ENVO:city	Titanium	NA	NA	NA	NA	NA	flores_restroom_surface_biogeography	restroom surface biogeography	256318	CCME	NA	NA	NA	1335	"Genomic DNA was extracted from the swabs and filters using the MO BIO PowerSoil DNA isolation kit following the manufacturer's protocol with the modifications of Fierer?et al.?[14]. A portion of the 16 S rRNA gene spanning the V1_V2 regions was amplified using the primer set (27 F/338R), PCR mixture conditions and thermal cycling conditions described in Fierer?et al.?[15]. PCR amplicons of triplicate reactions for each sample were pooled at approximately equal amounts and pyrosequenced at 454 Life Sciences (Branford, CT, USA) on their GS Junior system. A total of 337,333 high-quality partial 16 S rRNA gene sequences were obtained from 101 of the 120 surface samples collected, averaging approximately 3,340 sequences per sample (ranging from 513_6,771) (Table S1) in 4 GS Junior runs, with the best run containing 116,004 high-quality reads. An additional 16,416 sequences (ranging from 2161_5084 per sample) were generated for five of the six water samples collected for source tracking analysis. Each sample was amplified with a unique barcode to enable multiplexing in the GS Junior runs. The barcoded sequencing reads can be separated by data analysis software providing high confidence in assigning sequencing read to each sample. Sequence data generated as part of this study is available upon request by contacting the corresponding author."	NA	NA	y	NA	NA	ENVO:surface	16S rRNA	Ekeley	NA	V2	CCME	NA	Sink floor	NA	NA	NA	40.01499	NA	NA	Floor between sink and door
EKAM10.489518	GAATGATGAGTG	YATGCTGCCTCCCGTAGGAGT	jesse.stombaugh@colorado.edu	NA	restroom surface	Floor_between_sink_and_door	NA	NA	NA	n	12/13/10	restroom surface metagenome	NA	GFL0IUQ	NA	mimarks-survey	NA	NA	NA	NA	0	NA	NA	1624.1	10-Nov	Male	0	NA	NA	NA	NA	NA	NA	NA	NA	NA	NA	"1,swab"	A	NA	Microbial biogeography of public restroom surfaces	NA	NA	-105.27055	y	NA	CCME	NA	NA	NA	"We spend the majority of our lives indoors where we are constantly exposed to bacteria residing on surfaces. However, the diversity of these surface-associated communities is largely unknown. We explored the biogeographical patterns exhibited by bacteria across ten surfaces within each of twelve public restrooms. Using high-throughput barcoded pyrosequencing of the 16S rRNA gene, we identified 19 bacterial phyla across all surfaces. Most sequences belonged to four phyla: Actinobacteria, Bacteriodetes, Firmicutes and Proteobacteria. The communities clustered into three general categories: those found on surfaces associated with toilets, those on the restroom floor, and those found on surfaces routinely touched with hands. On toilet surfaces, gut-associated taxa were more prevalent, suggesting fecal contamination of these surfaces. Floor surfaces were the most diverse of all communities and contained several taxa commonly found in soils. Skin-associated bacteria, especially the Propionibacteriaceae, dominated surfaces routinely touched with our hands. Certain taxa were more common in female than in male restrooms as vagina-associated Lactobacillaceae were widely distributed in female restrooms, likely from urine contamination. Use of the SourceTracker algorithm confirmed many of our taxonomic observations as human skin was the primary source of bacteria on restroom surfaces. Overall, these results demonstrate that restroom surfaces host relatively diverse microbial communities dominated by human-associated bacteria with clear linkages between communities on or in different body sites and those communities found on restroom surfaces. More generally, this work is relevant to the public health field as we show that human associated microbes are commonly found on restroom surfaces suggesting that bacterial pathogens could readily be transmitted between individuals by the touching of surfaces. Furthermore, we demonstrate that we can use high-throughput analyses of bacterial communities to determine sources of bacteria on indoor surfaces, an approach which could be used to track pathogen transmission and test the efficacy of hygiene practices."	ENVO:building	TCAG	Microbial biogeography of public restroom surfaces	NA	0	"Ten surfaces (door handles into and out of the restroom, handles into and out of a restroom stall, faucet handles, soap dispenser, toilet seat, toilet flush handle, floor around the toilet and floor around the sink) in six male and six female restrooms evenly distributed across two buildings on the University of Colorado at Boulder campus were sampled on a single day in November 2010. Surfaces where sampled using sterile, cotton-tipped swabs as described previously?[14],?[15]. As the 12 restrooms were nearly identical in design, we were able to swab the same area at each location between restrooms. In order to characterize tap water communities as a potential source of bacteria, 1 L of faucet water from six of the restrooms (each building having the same water source for each restroom sampled) was collected and filtered through 0.2 ?m bottle top filters (Nalgene, Rochester, NY, USA)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	22132229	NA	NA	NA	Gilberto Flores	CCME	GAZ:United States of America	NA	NA	0	NA	NA	NA	NA	n	NA	Noah Fierer	Flores_restroom_surface_biogeography	pyrosequencing	NA	NA	NA	restroom surface metagenome	1	ENVO:city	Titanium	NA	NA	NA	NA	NA	flores_restroom_surface_biogeography	restroom surface biogeography	256318	CCME	NA	NA	NA	1335	"Genomic DNA was extracted from the swabs and filters using the MO BIO PowerSoil DNA isolation kit following the manufacturer's protocol with the modifications of Fierer?et al.?[14]. A portion of the 16 S rRNA gene spanning the V1_V2 regions was amplified using the primer set (27 F/338R), PCR mixture conditions and thermal cycling conditions described in Fierer?et al.?[15]. PCR amplicons of triplicate reactions for each sample were pooled at approximately equal amounts and pyrosequenced at 454 Life Sciences (Branford, CT, USA) on their GS Junior system. A total of 337,333 high-quality partial 16 S rRNA gene sequences were obtained from 101 of the 120 surface samples collected, averaging approximately 3,340 sequences per sample (ranging from 513_6,771) (Table S1) in 4 GS Junior runs, with the best run containing 116,004 high-quality reads. An additional 16,416 sequences (ranging from 2161_5084 per sample) were generated for five of the six water samples collected for source tracking analysis. Each sample was amplified with a unique barcode to enable multiplexing in the GS Junior runs. The barcoded sequencing reads can be separated by data analysis software providing high confidence in assigning sequencing read to each sample. Sequence data generated as part of this study is available upon request by contacting the corresponding author."	NA	NA	y	NA	NA	ENVO:surface	16S rRNA	Ekeley	NA	V2	CCME	NA	Sink floor	NA	NA	NA	40.01499	NA	NA	Floor between sink and door
EKBF10.489552	GAGTAGCTCGTG	YATGCTGCCTCCCGTAGGAGT	jesse.stombaugh@colorado.edu	NA	restroom surface	Floor_between_sink_and_door	NA	NA	NA	n	12/13/10	restroom surface metagenome	NA	GFL0IUQ	NA	mimarks-survey	NA	NA	NA	NA	0	NA	NA	1624.1	10-Nov	Female	0	NA	NA	NA	NA	NA	NA	NA	NA	NA	NA	"1,swab"	B	NA	Microbial biogeography of public restroom surfaces	NA	NA	-105.27055	y	NA	CCME	NA	NA	NA	"We spend the majority of our lives indoors where we are constantly exposed to bacteria residing on surfaces. However, the diversity of these surface-associated communities is largely unknown. We explored the biogeographical patterns exhibited by bacteria across ten surfaces within each of twelve public restrooms. Using high-throughput barcoded pyrosequencing of the 16S rRNA gene, we identified 19 bacterial phyla across all surfaces. Most sequences belonged to four phyla: Actinobacteria, Bacteriodetes, Firmicutes and Proteobacteria. The communities clustered into three general categories: those found on surfaces associated with toilets, those on the restroom floor, and those found on surfaces routinely touched with hands. On toilet surfaces, gut-associated taxa were more prevalent, suggesting fecal contamination of these surfaces. Floor surfaces were the most diverse of all communities and contained several taxa commonly found in soils. Skin-associated bacteria, especially the Propionibacteriaceae, dominated surfaces routinely touched with our hands. Certain taxa were more common in female than in male restrooms as vagina-associated Lactobacillaceae were widely distributed in female restrooms, likely from urine contamination. Use of the SourceTracker algorithm confirmed many of our taxonomic observations as human skin was the primary source of bacteria on restroom surfaces. Overall, these results demonstrate that restroom surfaces host relatively diverse microbial communities dominated by human-associated bacteria with clear linkages between communities on or in different body sites and those communities found on restroom surfaces. More generally, this work is relevant to the public health field as we show that human associated microbes are commonly found on restroom surfaces suggesting that bacterial pathogens could readily be transmitted between individuals by the touching of surfaces. Furthermore, we demonstrate that we can use high-throughput analyses of bacterial communities to determine sources of bacteria on indoor surfaces, an approach which could be used to track pathogen transmission and test the efficacy of hygiene practices."	ENVO:building	TCAG	Microbial biogeography of public restroom surfaces	NA	0	"Ten surfaces (door handles into and out of the restroom, handles into and out of a restroom stall, faucet handles, soap dispenser, toilet seat, toilet flush handle, floor around the toilet and floor around the sink) in six male and six female restrooms evenly distributed across two buildings on the University of Colorado at Boulder campus were sampled on a single day in November 2010. Surfaces where sampled using sterile, cotton-tipped swabs as described previously?[14],?[15]. As the 12 restrooms were nearly identical in design, we were able to swab the same area at each location between restrooms. In order to characterize tap water communities as a potential source of bacteria, 1 L of faucet water from six of the restrooms (each building having the same water source for each restroom sampled) was collected and filtered through 0.2 ?m bottle top filters (Nalgene, Rochester, NY, USA)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	22132229	NA	NA	NA	Gilberto Flores	CCME	GAZ:United States of America	NA	NA	0	NA	NA	NA	NA	n	NA	Noah Fierer	Flores_restroom_surface_biogeography	pyrosequencing	NA	NA	NA	restroom surface metagenome	2	ENVO:city	Titanium	NA	NA	NA	NA	NA	flores_restroom_surface_biogeography	restroom surface biogeography	256318	CCME	NA	NA	NA	1335	"Genomic DNA was extracted from the swabs and filters using the MO BIO PowerSoil DNA isolation kit following the manufacturer's protocol with the modifications of Fierer?et al.?[14]. A portion of the 16 S rRNA gene spanning the V1_V2 regions was amplified using the primer set (27 F/338R), PCR mixture conditions and thermal cycling conditions described in Fierer?et al.?[15]. PCR amplicons of triplicate reactions for each sample were pooled at approximately equal amounts and pyrosequenced at 454 Life Sciences (Branford, CT, USA) on their GS Junior system. A total of 337,333 high-quality partial 16 S rRNA gene sequences were obtained from 101 of the 120 surface samples collected, averaging approximately 3,340 sequences per sample (ranging from 513_6,771) (Table S1) in 4 GS Junior runs, with the best run containing 116,004 high-quality reads. An additional 16,416 sequences (ranging from 2161_5084 per sample) were generated for five of the six water samples collected for source tracking analysis. Each sample was amplified with a unique barcode to enable multiplexing in the GS Junior runs. The barcoded sequencing reads can be separated by data analysis software providing high confidence in assigning sequencing read to each sample. Sequence data generated as part of this study is available upon request by contacting the corresponding author."	NA	NA	y	NA	NA	ENVO:surface	16S rRNA	Ekeley	NA	V2	CCME	NA	Sink floor	NA	NA	NA	40.01499	NA	NA	Floor between sink and door
EKBM10.489477	GACTTCAGTGTG	YATGCTGCCTCCCGTAGGAGT	jesse.stombaugh@colorado.edu	NA	restroom surface	Floor_between_sink_and_door	NA	NA	NA	n	12/13/10	restroom surface metagenome	NA	GFL0IUQ	NA	mimarks-survey	NA	NA	NA	NA	0	NA	NA	1624.1	10-Nov	Male	0	NA	NA	NA	NA	NA	NA	NA	NA	NA	NA	"1,swab"	B	NA	Microbial biogeography of public restroom surfaces	NA	NA	-105.27055	y	NA	CCME	NA	NA	NA	"We spend the majority of our lives indoors where we are constantly exposed to bacteria residing on surfaces. However, the diversity of these surface-associated communities is largely unknown. We explored the biogeographical patterns exhibited by bacteria across ten surfaces within each of twelve public restrooms. Using high-throughput barcoded pyrosequencing of the 16S rRNA gene, we identified 19 bacterial phyla across all surfaces. Most sequences belonged to four phyla: Actinobacteria, Bacteriodetes, Firmicutes and Proteobacteria. The communities clustered into three general categories: those found on surfaces associated with toilets, those on the restroom floor, and those found on surfaces routinely touched with hands. On toilet surfaces, gut-associated taxa were more prevalent, suggesting fecal contamination of these surfaces. Floor surfaces were the most diverse of all communities and contained several taxa commonly found in soils. Skin-associated bacteria, especially the Propionibacteriaceae, dominated surfaces routinely touched with our hands. Certain taxa were more common in female than in male restrooms as vagina-associated Lactobacillaceae were widely distributed in female restrooms, likely from urine contamination. Use of the SourceTracker algorithm confirmed many of our taxonomic observations as human skin was the primary source of bacteria on restroom surfaces. Overall, these results demonstrate that restroom surfaces host relatively diverse microbial communities dominated by human-associated bacteria with clear linkages between communities on or in different body sites and those communities found on restroom surfaces. More generally, this work is relevant to the public health field as we show that human associated microbes are commonly found on restroom surfaces suggesting that bacterial pathogens could readily be transmitted between individuals by the touching of surfaces. Furthermore, we demonstrate that we can use high-throughput analyses of bacterial communities to determine sources of bacteria on indoor surfaces, an approach which could be used to track pathogen transmission and test the efficacy of hygiene practices."	ENVO:building	TCAG	Microbial biogeography of public restroom surfaces	NA	0	"Ten surfaces (door handles into and out of the restroom, handles into and out of a restroom stall, faucet handles, soap dispenser, toilet seat, toilet flush handle, floor around the toilet and floor around the sink) in six male and six female restrooms evenly distributed across two buildings on the University of Colorado at Boulder campus were sampled on a single day in November 2010. Surfaces where sampled using sterile, cotton-tipped swabs as described previously?[14],?[15]. As the 12 restrooms were nearly identical in design, we were able to swab the same area at each location between restrooms. In order to characterize tap water communities as a potential source of bacteria, 1 L of faucet water from six of the restrooms (each building having the same water source for each restroom sampled) was collected and filtered through 0.2 ?m bottle top filters (Nalgene, Rochester, NY, USA)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	22132229	NA	NA	NA	Gilberto Flores	CCME	GAZ:United States of America	NA	NA	0	NA	NA	NA	NA	n	NA	Noah Fierer	Flores_restroom_surface_biogeography	pyrosequencing	NA	NA	NA	restroom surface metagenome	2	ENVO:city	Titanium	NA	NA	NA	NA	NA	flores_restroom_surface_biogeography	restroom surface biogeography	256318	CCME	NA	NA	NA	1335	"Genomic DNA was extracted from the swabs and filters using the MO BIO PowerSoil DNA isolation kit following the manufacturer's protocol with the modifications of Fierer?et al.?[14]. A portion of the 16 S rRNA gene spanning the V1_V2 regions was amplified using the primer set (27 F/338R), PCR mixture conditions and thermal cycling conditions described in Fierer?et al.?[15]. PCR amplicons of triplicate reactions for each sample were pooled at approximately equal amounts and pyrosequenced at 454 Life Sciences (Branford, CT, USA) on their GS Junior system. A total of 337,333 high-quality partial 16 S rRNA gene sequences were obtained from 101 of the 120 surface samples collected, averaging approximately 3,340 sequences per sample (ranging from 513_6,771) (Table S1) in 4 GS Junior runs, with the best run containing 116,004 high-quality reads. An additional 16,416 sequences (ranging from 2161_5084 per sample) were generated for five of the six water samples collected for source tracking analysis. Each sample was amplified with a unique barcode to enable multiplexing in the GS Junior runs. The barcoded sequencing reads can be separated by data analysis software providing high confidence in assigning sequencing read to each sample. Sequence data generated as part of this study is available upon request by contacting the corresponding author."	NA	NA	y	NA	NA	ENVO:surface	16S rRNA	Ekeley	NA	V2	CCME	NA	Sink floor	NA	NA	NA	40.01499	NA	NA	Floor between sink and door
EKCF10.489541	GATTAGCACTCT	YATGCTGCCTCCCGTAGGAGT	jesse.stombaugh@colorado.edu	NA	restroom surface	Floor_between_sink_and_door	NA	NA	NA	n	12/13/10	restroom surface metagenome	NA	GFL0IUQ	NA	mimarks-survey	NA	NA	NA	NA	0	NA	NA	1624.1	10-Nov	Female	0	NA	NA	NA	NA	NA	NA	NA	NA	NA	NA	"1,swab"	C	NA	Microbial biogeography of public restroom surfaces	NA	NA	-105.27055	y	NA	CCME	NA	NA	NA	"We spend the majority of our lives indoors where we are constantly exposed to bacteria residing on surfaces. However, the diversity of these surface-associated communities is largely unknown. We explored the biogeographical patterns exhibited by bacteria across ten surfaces within each of twelve public restrooms. Using high-throughput barcoded pyrosequencing of the 16S rRNA gene, we identified 19 bacterial phyla across all surfaces. Most sequences belonged to four phyla: Actinobacteria, Bacteriodetes, Firmicutes and Proteobacteria. The communities clustered into three general categories: those found on surfaces associated with toilets, those on the restroom floor, and those found on surfaces routinely touched with hands. On toilet surfaces, gut-associated taxa were more prevalent, suggesting fecal contamination of these surfaces. Floor surfaces were the most diverse of all communities and contained several taxa commonly found in soils. Skin-associated bacteria, especially the Propionibacteriaceae, dominated surfaces routinely touched with our hands. Certain taxa were more common in female than in male restrooms as vagina-associated Lactobacillaceae were widely distributed in female restrooms, likely from urine contamination. Use of the SourceTracker algorithm confirmed many of our taxonomic observations as human skin was the primary source of bacteria on restroom surfaces. Overall, these results demonstrate that restroom surfaces host relatively diverse microbial communities dominated by human-associated bacteria with clear linkages between communities on or in different body sites and those communities found on restroom surfaces. More generally, this work is relevant to the public health field as we show that human associated microbes are commonly found on restroom surfaces suggesting that bacterial pathogens could readily be transmitted between individuals by the touching of surfaces. Furthermore, we demonstrate that we can use high-throughput analyses of bacterial communities to determine sources of bacteria on indoor surfaces, an approach which could be used to track pathogen transmission and test the efficacy of hygiene practices."	ENVO:building	TCAG	Microbial biogeography of public restroom surfaces	NA	0	"Ten surfaces (door handles into and out of the restroom, handles into and out of a restroom stall, faucet handles, soap dispenser, toilet seat, toilet flush handle, floor around the toilet and floor around the sink) in six male and six female restrooms evenly distributed across two buildings on the University of Colorado at Boulder campus were sampled on a single day in November 2010. Surfaces where sampled using sterile, cotton-tipped swabs as described previously?[14],?[15]. As the 12 restrooms were nearly identical in design, we were able to swab the same area at each location between restrooms. In order to characterize tap water communities as a potential source of bacteria, 1 L of faucet water from six of the restrooms (each building having the same water source for each restroom sampled) was collected and filtered through 0.2 ?m bottle top filters (Nalgene, Rochester, NY, USA)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	22132229	NA	NA	NA	Gilberto Flores	CCME	GAZ:United States of America	NA	NA	0	NA	NA	NA	NA	n	NA	Noah Fierer	Flores_restroom_surface_biogeography	pyrosequencing	NA	NA	NA	restroom surface metagenome	3	ENVO:city	Titanium	NA	NA	NA	NA	NA	flores_restroom_surface_biogeography	restroom surface biogeography	256318	CCME	NA	NA	NA	1335	"Genomic DNA was extracted from the swabs and filters using the MO BIO PowerSoil DNA isolation kit following the manufacturer's protocol with the modifications of Fierer?et al.?[14]. A portion of the 16 S rRNA gene spanning the V1_V2 regions was amplified using the primer set (27 F/338R), PCR mixture conditions and thermal cycling conditions described in Fierer?et al.?[15]. PCR amplicons of triplicate reactions for each sample were pooled at approximately equal amounts and pyrosequenced at 454 Life Sciences (Branford, CT, USA) on their GS Junior system. A total of 337,333 high-quality partial 16 S rRNA gene sequences were obtained from 101 of the 120 surface samples collected, averaging approximately 3,340 sequences per sample (ranging from 513_6,771) (Table S1) in 4 GS Junior runs, with the best run containing 116,004 high-quality reads. An additional 16,416 sequences (ranging from 2161_5084 per sample) were generated for five of the six water samples collected for source tracking analysis. Each sample was amplified with a unique barcode to enable multiplexing in the GS Junior runs. The barcoded sequencing reads can be separated by data analysis software providing high confidence in assigning sequencing read to each sample. Sequence data generated as part of this study is available upon request by contacting the corresponding author."	NA	NA	y	NA	NA	ENVO:surface	16S rRNA	Ekeley	NA	V2	CCME	NA	Sink floor	NA	NA	NA	40.01499	NA	NA	Floor between sink and door
EKCM10.489502	GATCCGACACTA	YATGCTGCCTCCCGTAGGAGT	jesse.stombaugh@colorado.edu	NA	restroom surface	Floor_between_sink_and_door	NA	NA	NA	n	12/13/10	restroom surface metagenome	NA	GFL0IUQ	NA	mimarks-survey	NA	NA	NA	NA	0	NA	NA	1624.1	10-Nov	Male	0	NA	NA	NA	NA	NA	NA	NA	NA	NA	NA	"1,swab"	C	NA	Microbial biogeography of public restroom surfaces	NA	NA	-105.27055	y	NA	CCME	NA	NA	NA	"We spend the majority of our lives indoors where we are constantly exposed to bacteria residing on surfaces. However, the diversity of these surface-associated communities is largely unknown. We explored the biogeographical patterns exhibited by bacteria across ten surfaces within each of twelve public restrooms. Using high-throughput barcoded pyrosequencing of the 16S rRNA gene, we identified 19 bacterial phyla across all surfaces. Most sequences belonged to four phyla: Actinobacteria, Bacteriodetes, Firmicutes and Proteobacteria. The communities clustered into three general categories: those found on surfaces associated with toilets, those on the restroom floor, and those found on surfaces routinely touched with hands. On toilet surfaces, gut-associated taxa were more prevalent, suggesting fecal contamination of these surfaces. Floor surfaces were the most diverse of all communities and contained several taxa commonly found in soils. Skin-associated bacteria, especially the Propionibacteriaceae, dominated surfaces routinely touched with our hands. Certain taxa were more common in female than in male restrooms as vagina-associated Lactobacillaceae were widely distributed in female restrooms, likely from urine contamination. Use of the SourceTracker algorithm confirmed many of our taxonomic observations as human skin was the primary source of bacteria on restroom surfaces. Overall, these results demonstrate that restroom surfaces host relatively diverse microbial communities dominated by human-associated bacteria with clear linkages between communities on or in different body sites and those communities found on restroom surfaces. More generally, this work is relevant to the public health field as we show that human associated microbes are commonly found on restroom surfaces suggesting that bacterial pathogens could readily be transmitted between individuals by the touching of surfaces. Furthermore, we demonstrate that we can use high-throughput analyses of bacterial communities to determine sources of bacteria on indoor surfaces, an approach which could be used to track pathogen transmission and test the efficacy of hygiene practices."	ENVO:building	TCAG	Microbial biogeography of public restroom surfaces	NA	0	"Ten surfaces (door handles into and out of the restroom, handles into and out of a restroom stall, faucet handles, soap dispenser, toilet seat, toilet flush handle, floor around the toilet and floor around the sink) in six male and six female restrooms evenly distributed across two buildings on the University of Colorado at Boulder campus were sampled on a single day in November 2010. Surfaces where sampled using sterile, cotton-tipped swabs as described previously?[14],?[15]. As the 12 restrooms were nearly identical in design, we were able to swab the same area at each location between restrooms. In order to characterize tap water communities as a potential source of bacteria, 1 L of faucet water from six of the restrooms (each building having the same water source for each restroom sampled) was collected and filtered through 0.2 ?m bottle top filters (Nalgene, Rochester, NY, USA)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	22132229	NA	NA	NA	Gilberto Flores	CCME	GAZ:United States of America	NA	NA	0	NA	NA	NA	NA	n	NA	Noah Fierer	Flores_restroom_surface_biogeography	pyrosequencing	NA	NA	NA	restroom surface metagenome	3	ENVO:city	Titanium	NA	NA	NA	NA	NA	flores_restroom_surface_biogeography	restroom surface biogeography	256318	CCME	NA	NA	NA	1335	"Genomic DNA was extracted from the swabs and filters using the MO BIO PowerSoil DNA isolation kit following the manufacturer's protocol with the modifications of Fierer?et al.?[14]. A portion of the 16 S rRNA gene spanning the V1_V2 regions was amplified using the primer set (27 F/338R), PCR mixture conditions and thermal cycling conditions described in Fierer?et al.?[15]. PCR amplicons of triplicate reactions for each sample were pooled at approximately equal amounts and pyrosequenced at 454 Life Sciences (Branford, CT, USA) on their GS Junior system. A total of 337,333 high-quality partial 16 S rRNA gene sequences were obtained from 101 of the 120 surface samples collected, averaging approximately 3,340 sequences per sample (ranging from 513_6,771) (Table S1) in 4 GS Junior runs, with the best run containing 116,004 high-quality reads. An additional 16,416 sequences (ranging from 2161_5084 per sample) were generated for five of the six water samples collected for source tracking analysis. Each sample was amplified with a unique barcode to enable multiplexing in the GS Junior runs. The barcoded sequencing reads can be separated by data analysis software providing high confidence in assigning sequencing read to each sample. Sequence data generated as part of this study is available upon request by contacting the corresponding author."	NA	NA	y	NA	NA	ENVO:surface	16S rRNA	Ekeley	NA	V2	CCME	NA	Sink floor	NA	NA	NA	40.01499	NA	NA	Floor between sink and door
PTAF10.489458	GCGACTTGTGTA	YATGCTGCCTCCCGTAGGAGT	jesse.stombaugh@colorado.edu	NA	restroom surface	Floor_between_sink_and_door	NA	NA	NA	n	12/13/10	restroom surface metagenome	NA	GFL0IUQ	NA	mimarks-survey	NA	NA	NA	NA	0	NA	NA	1624.1	10-Nov	Female	0	NA	NA	NA	NA	NA	NA	NA	NA	NA	NA	"1,swab"	A	NA	Microbial biogeography of public restroom surfaces	NA	NA	-105.27055	y	NA	CCME	NA	NA	NA	"We spend the majority of our lives indoors where we are constantly exposed to bacteria residing on surfaces. However, the diversity of these surface-associated communities is largely unknown. We explored the biogeographical patterns exhibited by bacteria across ten surfaces within each of twelve public restrooms. Using high-throughput barcoded pyrosequencing of the 16S rRNA gene, we identified 19 bacterial phyla across all surfaces. Most sequences belonged to four phyla: Actinobacteria, Bacteriodetes, Firmicutes and Proteobacteria. The communities clustered into three general categories: those found on surfaces associated with toilets, those on the restroom floor, and those found on surfaces routinely touched with hands. On toilet surfaces, gut-associated taxa were more prevalent, suggesting fecal contamination of these surfaces. Floor surfaces were the most diverse of all communities and contained several taxa commonly found in soils. Skin-associated bacteria, especially the Propionibacteriaceae, dominated surfaces routinely touched with our hands. Certain taxa were more common in female than in male restrooms as vagina-associated Lactobacillaceae were widely distributed in female restrooms, likely from urine contamination. Use of the SourceTracker algorithm confirmed many of our taxonomic observations as human skin was the primary source of bacteria on restroom surfaces. Overall, these results demonstrate that restroom surfaces host relatively diverse microbial communities dominated by human-associated bacteria with clear linkages between communities on or in different body sites and those communities found on restroom surfaces. More generally, this work is relevant to the public health field as we show that human associated microbes are commonly found on restroom surfaces suggesting that bacterial pathogens could readily be transmitted between individuals by the touching of surfaces. Furthermore, we demonstrate that we can use high-throughput analyses of bacterial communities to determine sources of bacteria on indoor surfaces, an approach which could be used to track pathogen transmission and test the efficacy of hygiene practices."	ENVO:building	TCAG	Microbial biogeography of public restroom surfaces	NA	0	"Ten surfaces (door handles into and out of the restroom, handles into and out of a restroom stall, faucet handles, soap dispenser, toilet seat, toilet flush handle, floor around the toilet and floor around the sink) in six male and six female restrooms evenly distributed across two buildings on the University of Colorado at Boulder campus were sampled on a single day in November 2010. Surfaces where sampled using sterile, cotton-tipped swabs as described previously?[14],?[15]. As the 12 restrooms were nearly identical in design, we were able to swab the same area at each location between restrooms. In order to characterize tap water communities as a potential source of bacteria, 1 L of faucet water from six of the restrooms (each building having the same water source for each restroom sampled) was collected and filtered through 0.2 ?m bottle top filters (Nalgene, Rochester, NY, USA)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	22132229	NA	NA	NA	Gilberto Flores	CCME	GAZ:United States of America	NA	NA	0	NA	NA	NA	NA	n	NA	Noah Fierer	Flores_restroom_surface_biogeography	pyrosequencing	NA	NA	NA	restroom surface metagenome	1	ENVO:city	Titanium	NA	NA	NA	NA	NA	flores_restroom_surface_biogeography	restroom surface biogeography	256318	CCME	NA	NA	NA	1335	"Genomic DNA was extracted from the swabs and filters using the MO BIO PowerSoil DNA isolation kit following the manufacturer's protocol with the modifications of Fierer?et al.?[14]. A portion of the 16 S rRNA gene spanning the V1_V2 regions was amplified using the primer set (27 F/338R), PCR mixture conditions and thermal cycling conditions described in Fierer?et al.?[15]. PCR amplicons of triplicate reactions for each sample were pooled at approximately equal amounts and pyrosequenced at 454 Life Sciences (Branford, CT, USA) on their GS Junior system. A total of 337,333 high-quality partial 16 S rRNA gene sequences were obtained from 101 of the 120 surface samples collected, averaging approximately 3,340 sequences per sample (ranging from 513_6,771) (Table S1) in 4 GS Junior runs, with the best run containing 116,004 high-quality reads. An additional 16,416 sequences (ranging from 2161_5084 per sample) were generated for five of the six water samples collected for source tracking analysis. Each sample was amplified with a unique barcode to enable multiplexing in the GS Junior runs. The barcoded sequencing reads can be separated by data analysis software providing high confidence in assigning sequencing read to each sample. Sequence data generated as part of this study is available upon request by contacting the corresponding author."	NA	NA	y	NA	NA	ENVO:surface	16S rRNA	Porter	NA	V2	CCME	NA	Sink floor	NA	NA	NA	40.01499	NA	NA	Floor between sink and door
PTAM10.489487	GCAGTATCACTG	YATGCTGCCTCCCGTAGGAGT	jesse.stombaugh@colorado.edu	NA	restroom surface	Floor_between_sink_and_door	NA	NA	NA	n	12/13/10	restroom surface metagenome	NA	GFL0IUQ	NA	mimarks-survey	NA	NA	NA	NA	0	NA	NA	1624.1	10-Nov	Male	0	NA	NA	NA	NA	NA	NA	NA	NA	NA	NA	"1,swab"	A	NA	Microbial biogeography of public restroom surfaces	NA	NA	-105.27055	y	NA	CCME	NA	NA	NA	"We spend the majority of our lives indoors where we are constantly exposed to bacteria residing on surfaces. However, the diversity of these surface-associated communities is largely unknown. We explored the biogeographical patterns exhibited by bacteria across ten surfaces within each of twelve public restrooms. Using high-throughput barcoded pyrosequencing of the 16S rRNA gene, we identified 19 bacterial phyla across all surfaces. Most sequences belonged to four phyla: Actinobacteria, Bacteriodetes, Firmicutes and Proteobacteria. The communities clustered into three general categories: those found on surfaces associated with toilets, those on the restroom floor, and those found on surfaces routinely touched with hands. On toilet surfaces, gut-associated taxa were more prevalent, suggesting fecal contamination of these surfaces. Floor surfaces were the most diverse of all communities and contained several taxa commonly found in soils. Skin-associated bacteria, especially the Propionibacteriaceae, dominated surfaces routinely touched with our hands. Certain taxa were more common in female than in male restrooms as vagina-associated Lactobacillaceae were widely distributed in female restrooms, likely from urine contamination. Use of the SourceTracker algorithm confirmed many of our taxonomic observations as human skin was the primary source of bacteria on restroom surfaces. Overall, these results demonstrate that restroom surfaces host relatively diverse microbial communities dominated by human-associated bacteria with clear linkages between communities on or in different body sites and those communities found on restroom surfaces. More generally, this work is relevant to the public health field as we show that human associated microbes are commonly found on restroom surfaces suggesting that bacterial pathogens could readily be transmitted between individuals by the touching of surfaces. Furthermore, we demonstrate that we can use high-throughput analyses of bacterial communities to determine sources of bacteria on indoor surfaces, an approach which could be used to track pathogen transmission and test the efficacy of hygiene practices."	ENVO:building	TCAG	Microbial biogeography of public restroom surfaces	NA	0	"Ten surfaces (door handles into and out of the restroom, handles into and out of a restroom stall, faucet handles, soap dispenser, toilet seat, toilet flush handle, floor around the toilet and floor around the sink) in six male and six female restrooms evenly distributed across two buildings on the University of Colorado at Boulder campus were sampled on a single day in November 2010. Surfaces where sampled using sterile, cotton-tipped swabs as described previously?[14],?[15]. As the 12 restrooms were nearly identical in design, we were able to swab the same area at each location between restrooms. In order to characterize tap water communities as a potential source of bacteria, 1 L of faucet water from six of the restrooms (each building having the same water source for each restroom sampled) was collected and filtered through 0.2 ?m bottle top filters (Nalgene, Rochester, NY, USA)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	22132229	NA	NA	NA	Gilberto Flores	CCME	GAZ:United States of America	NA	NA	0	NA	NA	NA	NA	n	NA	Noah Fierer	Flores_restroom_surface_biogeography	pyrosequencing	NA	NA	NA	restroom surface metagenome	1	ENVO:city	Titanium	NA	NA	NA	NA	NA	flores_restroom_surface_biogeography	restroom surface biogeography	256318	CCME	NA	NA	NA	1335	"Genomic DNA was extracted from the swabs and filters using the MO BIO PowerSoil DNA isolation kit following the manufacturer's protocol with the modifications of Fierer?et al.?[14]. A portion of the 16 S rRNA gene spanning the V1_V2 regions was amplified using the primer set (27 F/338R), PCR mixture conditions and thermal cycling conditions described in Fierer?et al.?[15]. PCR amplicons of triplicate reactions for each sample were pooled at approximately equal amounts and pyrosequenced at 454 Life Sciences (Branford, CT, USA) on their GS Junior system. A total of 337,333 high-quality partial 16 S rRNA gene sequences were obtained from 101 of the 120 surface samples collected, averaging approximately 3,340 sequences per sample (ranging from 513_6,771) (Table S1) in 4 GS Junior runs, with the best run containing 116,004 high-quality reads. An additional 16,416 sequences (ranging from 2161_5084 per sample) were generated for five of the six water samples collected for source tracking analysis. Each sample was amplified with a unique barcode to enable multiplexing in the GS Junior runs. The barcoded sequencing reads can be separated by data analysis software providing high confidence in assigning sequencing read to each sample. Sequence data generated as part of this study is available upon request by contacting the corresponding author."	NA	NA	y	NA	NA	ENVO:surface	16S rRNA	Porter	NA	V2	CCME	NA	Sink floor	NA	NA	NA	40.01499	NA	NA	Floor between sink and door
PTBF10.489566	AGTGTTCGATCG	YATGCTGCCTCCCGTAGGAGT	jesse.stombaugh@colorado.edu	NA	restroom surface	Floor_between_sink_and_door	NA	NA	NA	n	12/13/10	restroom surface metagenome	NA	GFL0IUQ	NA	mimarks-survey	NA	NA	NA	NA	0	NA	NA	1624.1	10-Nov	Female	0	NA	NA	NA	NA	NA	NA	NA	NA	NA	NA	"1,swab"	B	NA	Microbial biogeography of public restroom surfaces	NA	NA	-105.27055	y	NA	CCME	NA	NA	NA	"We spend the majority of our lives indoors where we are constantly exposed to bacteria residing on surfaces. However, the diversity of these surface-associated communities is largely unknown. We explored the biogeographical patterns exhibited by bacteria across ten surfaces within each of twelve public restrooms. Using high-throughput barcoded pyrosequencing of the 16S rRNA gene, we identified 19 bacterial phyla across all surfaces. Most sequences belonged to four phyla: Actinobacteria, Bacteriodetes, Firmicutes and Proteobacteria. The communities clustered into three general categories: those found on surfaces associated with toilets, those on the restroom floor, and those found on surfaces routinely touched with hands. On toilet surfaces, gut-associated taxa were more prevalent, suggesting fecal contamination of these surfaces. Floor surfaces were the most diverse of all communities and contained several taxa commonly found in soils. Skin-associated bacteria, especially the Propionibacteriaceae, dominated surfaces routinely touched with our hands. Certain taxa were more common in female than in male restrooms as vagina-associated Lactobacillaceae were widely distributed in female restrooms, likely from urine contamination. Use of the SourceTracker algorithm confirmed many of our taxonomic observations as human skin was the primary source of bacteria on restroom surfaces. Overall, these results demonstrate that restroom surfaces host relatively diverse microbial communities dominated by human-associated bacteria with clear linkages between communities on or in different body sites and those communities found on restroom surfaces. More generally, this work is relevant to the public health field as we show that human associated microbes are commonly found on restroom surfaces suggesting that bacterial pathogens could readily be transmitted between individuals by the touching of surfaces. Furthermore, we demonstrate that we can use high-throughput analyses of bacterial communities to determine sources of bacteria on indoor surfaces, an approach which could be used to track pathogen transmission and test the efficacy of hygiene practices."	ENVO:building	TCAG	Microbial biogeography of public restroom surfaces	NA	0	"Ten surfaces (door handles into and out of the restroom, handles into and out of a restroom stall, faucet handles, soap dispenser, toilet seat, toilet flush handle, floor around the toilet and floor around the sink) in six male and six female restrooms evenly distributed across two buildings on the University of Colorado at Boulder campus were sampled on a single day in November 2010. Surfaces where sampled using sterile, cotton-tipped swabs as described previously?[14],?[15]. As the 12 restrooms were nearly identical in design, we were able to swab the same area at each location between restrooms. In order to characterize tap water communities as a potential source of bacteria, 1 L of faucet water from six of the restrooms (each building having the same water source for each restroom sampled) was collected and filtered through 0.2 ?m bottle top filters (Nalgene, Rochester, NY, USA)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	22132229	NA	NA	NA	Gilberto Flores	CCME	GAZ:United States of America	NA	NA	0	NA	NA	NA	NA	n	NA	Noah Fierer	Flores_restroom_surface_biogeography	pyrosequencing	NA	NA	NA	restroom surface metagenome	2	ENVO:city	Titanium	NA	NA	NA	NA	NA	flores_restroom_surface_biogeography	restroom surface biogeography	256318	CCME	NA	NA	NA	1335	"Genomic DNA was extracted from the swabs and filters using the MO BIO PowerSoil DNA isolation kit following the manufacturer's protocol with the modifications of Fierer?et al.?[14]. A portion of the 16 S rRNA gene spanning the V1_V2 regions was amplified using the primer set (27 F/338R), PCR mixture conditions and thermal cycling conditions described in Fierer?et al.?[15]. PCR amplicons of triplicate reactions for each sample were pooled at approximately equal amounts and pyrosequenced at 454 Life Sciences (Branford, CT, USA) on their GS Junior system. A total of 337,333 high-quality partial 16 S rRNA gene sequences were obtained from 101 of the 120 surface samples collected, averaging approximately 3,340 sequences per sample (ranging from 513_6,771) (Table S1) in 4 GS Junior runs, with the best run containing 116,004 high-quality reads. An additional 16,416 sequences (ranging from 2161_5084 per sample) were generated for five of the six water samples collected for source tracking analysis. Each sample was amplified with a unique barcode to enable multiplexing in the GS Junior runs. The barcoded sequencing reads can be separated by data analysis software providing high confidence in assigning sequencing read to each sample. Sequence data generated as part of this study is available upon request by contacting the corresponding author."	NA	NA	y	NA	NA	ENVO:surface	16S rRNA	Porter	NA	V2	CCME	NA	Sink floor	NA	NA	NA	40.01499	NA	NA	Floor between sink and door
PTBM10.489535	GCTATCACGAGT	YATGCTGCCTCCCGTAGGAGT	jesse.stombaugh@colorado.edu	NA	restroom surface	Floor_between_sink_and_door	NA	NA	NA	n	12/13/10	restroom surface metagenome	NA	GFL0IUQ	NA	mimarks-survey	NA	NA	NA	NA	0	NA	NA	1624.1	10-Nov	Male	0	NA	NA	NA	NA	NA	NA	NA	NA	NA	NA	"1,swab"	B	NA	Microbial biogeography of public restroom surfaces	NA	NA	-105.27055	y	NA	CCME	NA	NA	NA	"We spend the majority of our lives indoors where we are constantly exposed to bacteria residing on surfaces. However, the diversity of these surface-associated communities is largely unknown. We explored the biogeographical patterns exhibited by bacteria across ten surfaces within each of twelve public restrooms. Using high-throughput barcoded pyrosequencing of the 16S rRNA gene, we identified 19 bacterial phyla across all surfaces. Most sequences belonged to four phyla: Actinobacteria, Bacteriodetes, Firmicutes and Proteobacteria. The communities clustered into three general categories: those found on surfaces associated with toilets, those on the restroom floor, and those found on surfaces routinely touched with hands. On toilet surfaces, gut-associated taxa were more prevalent, suggesting fecal contamination of these surfaces. Floor surfaces were the most diverse of all communities and contained several taxa commonly found in soils. Skin-associated bacteria, especially the Propionibacteriaceae, dominated surfaces routinely touched with our hands. Certain taxa were more common in female than in male restrooms as vagina-associated Lactobacillaceae were widely distributed in female restrooms, likely from urine contamination. Use of the SourceTracker algorithm confirmed many of our taxonomic observations as human skin was the primary source of bacteria on restroom surfaces. Overall, these results demonstrate that restroom surfaces host relatively diverse microbial communities dominated by human-associated bacteria with clear linkages between communities on or in different body sites and those communities found on restroom surfaces. More generally, this work is relevant to the public health field as we show that human associated microbes are commonly found on restroom surfaces suggesting that bacterial pathogens could readily be transmitted between individuals by the touching of surfaces. Furthermore, we demonstrate that we can use high-throughput analyses of bacterial communities to determine sources of bacteria on indoor surfaces, an approach which could be used to track pathogen transmission and test the efficacy of hygiene practices."	ENVO:building	TCAG	Microbial biogeography of public restroom surfaces	NA	0	"Ten surfaces (door handles into and out of the restroom, handles into and out of a restroom stall, faucet handles, soap dispenser, toilet seat, toilet flush handle, floor around the toilet and floor around the sink) in six male and six female restrooms evenly distributed across two buildings on the University of Colorado at Boulder campus were sampled on a single day in November 2010. Surfaces where sampled using sterile, cotton-tipped swabs as described previously?[14],?[15]. As the 12 restrooms were nearly identical in design, we were able to swab the same area at each location between restrooms. In order to characterize tap water communities as a potential source of bacteria, 1 L of faucet water from six of the restrooms (each building having the same water source for each restroom sampled) was collected and filtered through 0.2 ?m bottle top filters (Nalgene, Rochester, NY, USA)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	22132229	NA	NA	NA	Gilberto Flores	CCME	GAZ:United States of America	NA	NA	0	NA	NA	NA	NA	n	NA	Noah Fierer	Flores_restroom_surface_biogeography	pyrosequencing	NA	NA	NA	restroom surface metagenome	2	ENVO:city	Titanium	NA	NA	NA	NA	NA	flores_restroom_surface_biogeography	restroom surface biogeography	256318	CCME	NA	NA	NA	1335	"Genomic DNA was extracted from the swabs and filters using the MO BIO PowerSoil DNA isolation kit following the manufacturer's protocol with the modifications of Fierer?et al.?[14]. A portion of the 16 S rRNA gene spanning the V1_V2 regions was amplified using the primer set (27 F/338R), PCR mixture conditions and thermal cycling conditions described in Fierer?et al.?[15]. PCR amplicons of triplicate reactions for each sample were pooled at approximately equal amounts and pyrosequenced at 454 Life Sciences (Branford, CT, USA) on their GS Junior system. A total of 337,333 high-quality partial 16 S rRNA gene sequences were obtained from 101 of the 120 surface samples collected, averaging approximately 3,340 sequences per sample (ranging from 513_6,771) (Table S1) in 4 GS Junior runs, with the best run containing 116,004 high-quality reads. An additional 16,416 sequences (ranging from 2161_5084 per sample) were generated for five of the six water samples collected for source tracking analysis. Each sample was amplified with a unique barcode to enable multiplexing in the GS Junior runs. The barcoded sequencing reads can be separated by data analysis software providing high confidence in assigning sequencing read to each sample. Sequence data generated as part of this study is available upon request by contacting the corresponding author."	NA	NA	y	NA	NA	ENVO:surface	16S rRNA	Porter	NA	V2	CCME	NA	Sink floor	NA	NA	NA	40.01499	NA	NA	Floor between sink and door
PTCF10.489534	ATCGCGGACGAT	YATGCTGCCTCCCGTAGGAGT	jesse.stombaugh@colorado.edu	NA	restroom surface	Floor_between_sink_and_door	NA	NA	NA	n	12/13/10	restroom surface metagenome	NA	GFL0IUQ	NA	mimarks-survey	NA	NA	NA	NA	0	NA	NA	1624.1	10-Nov	Female	0	NA	NA	NA	NA	NA	NA	NA	NA	NA	NA	"1,swab"	C	NA	Microbial biogeography of public restroom surfaces	NA	NA	-105.27055	y	NA	CCME	NA	NA	NA	"We spend the majority of our lives indoors where we are constantly exposed to bacteria residing on surfaces. However, the diversity of these surface-associated communities is largely unknown. We explored the biogeographical patterns exhibited by bacteria across ten surfaces within each of twelve public restrooms. Using high-throughput barcoded pyrosequencing of the 16S rRNA gene, we identified 19 bacterial phyla across all surfaces. Most sequences belonged to four phyla: Actinobacteria, Bacteriodetes, Firmicutes and Proteobacteria. The communities clustered into three general categories: those found on surfaces associated with toilets, those on the restroom floor, and those found on surfaces routinely touched with hands. On toilet surfaces, gut-associated taxa were more prevalent, suggesting fecal contamination of these surfaces. Floor surfaces were the most diverse of all communities and contained several taxa commonly found in soils. Skin-associated bacteria, especially the Propionibacteriaceae, dominated surfaces routinely touched with our hands. Certain taxa were more common in female than in male restrooms as vagina-associated Lactobacillaceae were widely distributed in female restrooms, likely from urine contamination. Use of the SourceTracker algorithm confirmed many of our taxonomic observations as human skin was the primary source of bacteria on restroom surfaces. Overall, these results demonstrate that restroom surfaces host relatively diverse microbial communities dominated by human-associated bacteria with clear linkages between communities on or in different body sites and those communities found on restroom surfaces. More generally, this work is relevant to the public health field as we show that human associated microbes are commonly found on restroom surfaces suggesting that bacterial pathogens could readily be transmitted between individuals by the touching of surfaces. Furthermore, we demonstrate that we can use high-throughput analyses of bacterial communities to determine sources of bacteria on indoor surfaces, an approach which could be used to track pathogen transmission and test the efficacy of hygiene practices."	ENVO:building	TCAG	Microbial biogeography of public restroom surfaces	NA	0	"Ten surfaces (door handles into and out of the restroom, handles into and out of a restroom stall, faucet handles, soap dispenser, toilet seat, toilet flush handle, floor around the toilet and floor around the sink) in six male and six female restrooms evenly distributed across two buildings on the University of Colorado at Boulder campus were sampled on a single day in November 2010. Surfaces where sampled using sterile, cotton-tipped swabs as described previously?[14],?[15]. As the 12 restrooms were nearly identical in design, we were able to swab the same area at each location between restrooms. In order to characterize tap water communities as a potential source of bacteria, 1 L of faucet water from six of the restrooms (each building having the same water source for each restroom sampled) was collected and filtered through 0.2 ?m bottle top filters (Nalgene, Rochester, NY, USA)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	22132229	NA	NA	NA	Gilberto Flores	CCME	GAZ:United States of America	NA	NA	0	NA	NA	NA	NA	n	NA	Noah Fierer	Flores_restroom_surface_biogeography	pyrosequencing	NA	NA	NA	restroom surface metagenome	3	ENVO:city	Titanium	NA	NA	NA	NA	NA	flores_restroom_surface_biogeography	restroom surface biogeography	256318	CCME	NA	NA	NA	1335	"Genomic DNA was extracted from the swabs and filters using the MO BIO PowerSoil DNA isolation kit following the manufacturer's protocol with the modifications of Fierer?et al.?[14]. A portion of the 16 S rRNA gene spanning the V1_V2 regions was amplified using the primer set (27 F/338R), PCR mixture conditions and thermal cycling conditions described in Fierer?et al.?[15]. PCR amplicons of triplicate reactions for each sample were pooled at approximately equal amounts and pyrosequenced at 454 Life Sciences (Branford, CT, USA) on their GS Junior system. A total of 337,333 high-quality partial 16 S rRNA gene sequences were obtained from 101 of the 120 surface samples collected, averaging approximately 3,340 sequences per sample (ranging from 513_6,771) (Table S1) in 4 GS Junior runs, with the best run containing 116,004 high-quality reads. An additional 16,416 sequences (ranging from 2161_5084 per sample) were generated for five of the six water samples collected for source tracking analysis. Each sample was amplified with a unique barcode to enable multiplexing in the GS Junior runs. The barcoded sequencing reads can be separated by data analysis software providing high confidence in assigning sequencing read to each sample. Sequence data generated as part of this study is available upon request by contacting the corresponding author."	NA	NA	y	NA	NA	ENVO:surface	16S rRNA	Porter	NA	V2	CCME	NA	Sink floor	NA	NA	NA	40.01499	NA	NA	Floor between sink and door
PTCM10.489512	ATAGCTCCATAC	YATGCTGCCTCCCGTAGGAGT	jesse.stombaugh@colorado.edu	NA	restroom surface	Floor_between_sink_and_door	NA	NA	NA	n	12/13/10	restroom surface metagenome	NA	GFL0IUQ	NA	mimarks-survey	NA	NA	NA	NA	0	NA	NA	1624.1	10-Nov	Male	0	NA	NA	NA	NA	NA	NA	NA	NA	NA	NA	"1,swab"	C	NA	Microbial biogeography of public restroom surfaces	NA	NA	-105.27055	y	NA	CCME	NA	NA	NA	"We spend the majority of our lives indoors where we are constantly exposed to bacteria residing on surfaces. However, the diversity of these surface-associated communities is largely unknown. We explored the biogeographical patterns exhibited by bacteria across ten surfaces within each of twelve public restrooms. Using high-throughput barcoded pyrosequencing of the 16S rRNA gene, we identified 19 bacterial phyla across all surfaces. Most sequences belonged to four phyla: Actinobacteria, Bacteriodetes, Firmicutes and Proteobacteria. The communities clustered into three general categories: those found on surfaces associated with toilets, those on the restroom floor, and those found on surfaces routinely touched with hands. On toilet surfaces, gut-associated taxa were more prevalent, suggesting fecal contamination of these surfaces. Floor surfaces were the most diverse of all communities and contained several taxa commonly found in soils. Skin-associated bacteria, especially the Propionibacteriaceae, dominated surfaces routinely touched with our hands. Certain taxa were more common in female than in male restrooms as vagina-associated Lactobacillaceae were widely distributed in female restrooms, likely from urine contamination. Use of the SourceTracker algorithm confirmed many of our taxonomic observations as human skin was the primary source of bacteria on restroom surfaces. Overall, these results demonstrate that restroom surfaces host relatively diverse microbial communities dominated by human-associated bacteria with clear linkages between communities on or in different body sites and those communities found on restroom surfaces. More generally, this work is relevant to the public health field as we show that human associated microbes are commonly found on restroom surfaces suggesting that bacterial pathogens could readily be transmitted between individuals by the touching of surfaces. Furthermore, we demonstrate that we can use high-throughput analyses of bacterial communities to determine sources of bacteria on indoor surfaces, an approach which could be used to track pathogen transmission and test the efficacy of hygiene practices."	ENVO:building	TCAG	Microbial biogeography of public restroom surfaces	NA	0	"Ten surfaces (door handles into and out of the restroom, handles into and out of a restroom stall, faucet handles, soap dispenser, toilet seat, toilet flush handle, floor around the toilet and floor around the sink) in six male and six female restrooms evenly distributed across two buildings on the University of Colorado at Boulder campus were sampled on a single day in November 2010. Surfaces where sampled using sterile, cotton-tipped swabs as described previously?[14],?[15]. As the 12 restrooms were nearly identical in design, we were able to swab the same area at each location between restrooms. In order to characterize tap water communities as a potential source of bacteria, 1 L of faucet water from six of the restrooms (each building having the same water source for each restroom sampled) was collected and filtered through 0.2 ?m bottle top filters (Nalgene, Rochester, NY, USA)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	22132229	NA	NA	NA	Gilberto Flores	CCME	GAZ:United States of America	NA	NA	0	NA	NA	NA	NA	n	NA	Noah Fierer	Flores_restroom_surface_biogeography	pyrosequencing	NA	NA	NA	restroom surface metagenome	3	ENVO:city	Titanium	NA	NA	NA	NA	NA	flores_restroom_surface_biogeography	restroom surface biogeography	256318	CCME	NA	NA	NA	1335	"Genomic DNA was extracted from the swabs and filters using the MO BIO PowerSoil DNA isolation kit following the manufacturer's protocol with the modifications of Fierer?et al.?[14]. A portion of the 16 S rRNA gene spanning the V1_V2 regions was amplified using the primer set (27 F/338R), PCR mixture conditions and thermal cycling conditions described in Fierer?et al.?[15]. PCR amplicons of triplicate reactions for each sample were pooled at approximately equal amounts and pyrosequenced at 454 Life Sciences (Branford, CT, USA) on their GS Junior system. A total of 337,333 high-quality partial 16 S rRNA gene sequences were obtained from 101 of the 120 surface samples collected, averaging approximately 3,340 sequences per sample (ranging from 513_6,771) (Table S1) in 4 GS Junior runs, with the best run containing 116,004 high-quality reads. An additional 16,416 sequences (ranging from 2161_5084 per sample) were generated for five of the six water samples collected for source tracking analysis. Each sample was amplified with a unique barcode to enable multiplexing in the GS Junior runs. The barcoded sequencing reads can be separated by data analysis software providing high confidence in assigning sequencing read to each sample. Sequence data generated as part of this study is available upon request by contacting the corresponding author."	NA	NA	y	NA	NA	ENVO:surface	16S rRNA	Porter	NA	V2	CCME	NA	Sink floor	NA	NA	NA	40.01499	NA	NA	Floor between sink and door
EKAF7.489567	GACCGAGCTATG	YATGCTGCCTCCCGTAGGAGT	jesse.stombaugh@colorado.edu	NA	restroom surface	Floor_in_front_of_toilet	NA	NA	NA	n	12/13/10	restroom surface metagenome	NA	GFL0IUQ	NA	mimarks-survey	NA	NA	NA	NA	0	NA	NA	1624.1	10-Nov	Female	0	NA	NA	NA	NA	NA	NA	NA	NA	NA	NA	"1,swab"	A	NA	Microbial biogeography of public restroom surfaces	NA	NA	-105.27055	y	NA	CCME	NA	NA	NA	"We spend the majority of our lives indoors where we are constantly exposed to bacteria residing on surfaces. However, the diversity of these surface-associated communities is largely unknown. We explored the biogeographical patterns exhibited by bacteria across ten surfaces within each of twelve public restrooms. Using high-throughput barcoded pyrosequencing of the 16S rRNA gene, we identified 19 bacterial phyla across all surfaces. Most sequences belonged to four phyla: Actinobacteria, Bacteriodetes, Firmicutes and Proteobacteria. The communities clustered into three general categories: those found on surfaces associated with toilets, those on the restroom floor, and those found on surfaces routinely touched with hands. On toilet surfaces, gut-associated taxa were more prevalent, suggesting fecal contamination of these surfaces. Floor surfaces were the most diverse of all communities and contained several taxa commonly found in soils. Skin-associated bacteria, especially the Propionibacteriaceae, dominated surfaces routinely touched with our hands. Certain taxa were more common in female than in male restrooms as vagina-associated Lactobacillaceae were widely distributed in female restrooms, likely from urine contamination. Use of the SourceTracker algorithm confirmed many of our taxonomic observations as human skin was the primary source of bacteria on restroom surfaces. Overall, these results demonstrate that restroom surfaces host relatively diverse microbial communities dominated by human-associated bacteria with clear linkages between communities on or in different body sites and those communities found on restroom surfaces. More generally, this work is relevant to the public health field as we show that human associated microbes are commonly found on restroom surfaces suggesting that bacterial pathogens could readily be transmitted between individuals by the touching of surfaces. Furthermore, we demonstrate that we can use high-throughput analyses of bacterial communities to determine sources of bacteria on indoor surfaces, an approach which could be used to track pathogen transmission and test the efficacy of hygiene practices."	ENVO:building	TCAG	Microbial biogeography of public restroom surfaces	NA	0	"Ten surfaces (door handles into and out of the restroom, handles into and out of a restroom stall, faucet handles, soap dispenser, toilet seat, toilet flush handle, floor around the toilet and floor around the sink) in six male and six female restrooms evenly distributed across two buildings on the University of Colorado at Boulder campus were sampled on a single day in November 2010. Surfaces where sampled using sterile, cotton-tipped swabs as described previously?[14],?[15]. As the 12 restrooms were nearly identical in design, we were able to swab the same area at each location between restrooms. In order to characterize tap water communities as a potential source of bacteria, 1 L of faucet water from six of the restrooms (each building having the same water source for each restroom sampled) was collected and filtered through 0.2 ?m bottle top filters (Nalgene, Rochester, NY, USA)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	22132229	NA	NA	NA	Gilberto Flores	CCME	GAZ:United States of America	NA	NA	0	NA	NA	NA	NA	n	NA	Noah Fierer	Flores_restroom_surface_biogeography	pyrosequencing	NA	NA	NA	restroom surface metagenome	1	ENVO:city	Titanium	NA	NA	NA	NA	NA	flores_restroom_surface_biogeography	restroom surface biogeography	256318	CCME	NA	NA	NA	1335	"Genomic DNA was extracted from the swabs and filters using the MO BIO PowerSoil DNA isolation kit following the manufacturer's protocol with the modifications of Fierer?et al.?[14]. A portion of the 16 S rRNA gene spanning the V1_V2 regions was amplified using the primer set (27 F/338R), PCR mixture conditions and thermal cycling conditions described in Fierer?et al.?[15]. PCR amplicons of triplicate reactions for each sample were pooled at approximately equal amounts and pyrosequenced at 454 Life Sciences (Branford, CT, USA) on their GS Junior system. A total of 337,333 high-quality partial 16 S rRNA gene sequences were obtained from 101 of the 120 surface samples collected, averaging approximately 3,340 sequences per sample (ranging from 513_6,771) (Table S1) in 4 GS Junior runs, with the best run containing 116,004 high-quality reads. An additional 16,416 sequences (ranging from 2161_5084 per sample) were generated for five of the six water samples collected for source tracking analysis. Each sample was amplified with a unique barcode to enable multiplexing in the GS Junior runs. The barcoded sequencing reads can be separated by data analysis software providing high confidence in assigning sequencing read to each sample. Sequence data generated as part of this study is available upon request by contacting the corresponding author."	NA	NA	y	NA	NA	ENVO:surface	16S rRNA	Ekeley	NA	V2	CCME	NA	Toilet Floor	NA	NA	NA	40.01499	NA	NA	Floor in front of toilet
EKAM7.489572	GAAGAGTGATCA	YATGCTGCCTCCCGTAGGAGT	jesse.stombaugh@colorado.edu	NA	restroom surface	Floor_in_front_of_toilet	NA	NA	NA	n	12/13/10	restroom surface metagenome	NA	GFL0IUQ	NA	mimarks-survey	NA	NA	NA	NA	0	NA	NA	1624.1	10-Nov	Male	0	NA	NA	NA	NA	NA	NA	NA	NA	NA	NA	"1,swab"	A	NA	Microbial biogeography of public restroom surfaces	NA	NA	-105.27055	y	NA	CCME	NA	NA	NA	"We spend the majority of our lives indoors where we are constantly exposed to bacteria residing on surfaces. However, the diversity of these surface-associated communities is largely unknown. We explored the biogeographical patterns exhibited by bacteria across ten surfaces within each of twelve public restrooms. Using high-throughput barcoded pyrosequencing of the 16S rRNA gene, we identified 19 bacterial phyla across all surfaces. Most sequences belonged to four phyla: Actinobacteria, Bacteriodetes, Firmicutes and Proteobacteria. The communities clustered into three general categories: those found on surfaces associated with toilets, those on the restroom floor, and those found on surfaces routinely touched with hands. On toilet surfaces, gut-associated taxa were more prevalent, suggesting fecal contamination of these surfaces. Floor surfaces were the most diverse of all communities and contained several taxa commonly found in soils. Skin-associated bacteria, especially the Propionibacteriaceae, dominated surfaces routinely touched with our hands. Certain taxa were more common in female than in male restrooms as vagina-associated Lactobacillaceae were widely distributed in female restrooms, likely from urine contamination. Use of the SourceTracker algorithm confirmed many of our taxonomic observations as human skin was the primary source of bacteria on restroom surfaces. Overall, these results demonstrate that restroom surfaces host relatively diverse microbial communities dominated by human-associated bacteria with clear linkages between communities on or in different body sites and those communities found on restroom surfaces. More generally, this work is relevant to the public health field as we show that human associated microbes are commonly found on restroom surfaces suggesting that bacterial pathogens could readily be transmitted between individuals by the touching of surfaces. Furthermore, we demonstrate that we can use high-throughput analyses of bacterial communities to determine sources of bacteria on indoor surfaces, an approach which could be used to track pathogen transmission and test the efficacy of hygiene practices."	ENVO:building	TCAG	Microbial biogeography of public restroom surfaces	NA	0	"Ten surfaces (door handles into and out of the restroom, handles into and out of a restroom stall, faucet handles, soap dispenser, toilet seat, toilet flush handle, floor around the toilet and floor around the sink) in six male and six female restrooms evenly distributed across two buildings on the University of Colorado at Boulder campus were sampled on a single day in November 2010. Surfaces where sampled using sterile, cotton-tipped swabs as described previously?[14],?[15]. As the 12 restrooms were nearly identical in design, we were able to swab the same area at each location between restrooms. In order to characterize tap water communities as a potential source of bacteria, 1 L of faucet water from six of the restrooms (each building having the same water source for each restroom sampled) was collected and filtered through 0.2 ?m bottle top filters (Nalgene, Rochester, NY, USA)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	22132229	NA	NA	NA	Gilberto Flores	CCME	GAZ:United States of America	NA	NA	0	NA	NA	NA	NA	n	NA	Noah Fierer	Flores_restroom_surface_biogeography	pyrosequencing	NA	NA	NA	restroom surface metagenome	1	ENVO:city	Titanium	NA	NA	NA	NA	NA	flores_restroom_surface_biogeography	restroom surface biogeography	256318	CCME	NA	NA	NA	1335	"Genomic DNA was extracted from the swabs and filters using the MO BIO PowerSoil DNA isolation kit following the manufacturer's protocol with the modifications of Fierer?et al.?[14]. A portion of the 16 S rRNA gene spanning the V1_V2 regions was amplified using the primer set (27 F/338R), PCR mixture conditions and thermal cycling conditions described in Fierer?et al.?[15]. PCR amplicons of triplicate reactions for each sample were pooled at approximately equal amounts and pyrosequenced at 454 Life Sciences (Branford, CT, USA) on their GS Junior system. A total of 337,333 high-quality partial 16 S rRNA gene sequences were obtained from 101 of the 120 surface samples collected, averaging approximately 3,340 sequences per sample (ranging from 513_6,771) (Table S1) in 4 GS Junior runs, with the best run containing 116,004 high-quality reads. An additional 16,416 sequences (ranging from 2161_5084 per sample) were generated for five of the six water samples collected for source tracking analysis. Each sample was amplified with a unique barcode to enable multiplexing in the GS Junior runs. The barcoded sequencing reads can be separated by data analysis software providing high confidence in assigning sequencing read to each sample. Sequence data generated as part of this study is available upon request by contacting the corresponding author."	NA	NA	y	NA	NA	ENVO:surface	16S rRNA	Ekeley	NA	V2	CCME	NA	Toilet Floor	NA	NA	NA	40.01499	NA	NA	Floor in front of toilet
EKBF7.489524	GAGCATTCTCTA	YATGCTGCCTCCCGTAGGAGT	jesse.stombaugh@colorado.edu	NA	restroom surface	Floor_in_front_of_toilet	NA	NA	NA	n	12/13/10	restroom surface metagenome	NA	GFL0IUQ	NA	mimarks-survey	NA	NA	NA	NA	0	NA	NA	1624.1	10-Nov	Female	0	NA	NA	NA	NA	NA	NA	NA	NA	NA	NA	"1,swab"	B	NA	Microbial biogeography of public restroom surfaces	NA	NA	-105.27055	y	NA	CCME	NA	NA	NA	"We spend the majority of our lives indoors where we are constantly exposed to bacteria residing on surfaces. However, the diversity of these surface-associated communities is largely unknown. We explored the biogeographical patterns exhibited by bacteria across ten surfaces within each of twelve public restrooms. Using high-throughput barcoded pyrosequencing of the 16S rRNA gene, we identified 19 bacterial phyla across all surfaces. Most sequences belonged to four phyla: Actinobacteria, Bacteriodetes, Firmicutes and Proteobacteria. The communities clustered into three general categories: those found on surfaces associated with toilets, those on the restroom floor, and those found on surfaces routinely touched with hands. On toilet surfaces, gut-associated taxa were more prevalent, suggesting fecal contamination of these surfaces. Floor surfaces were the most diverse of all communities and contained several taxa commonly found in soils. Skin-associated bacteria, especially the Propionibacteriaceae, dominated surfaces routinely touched with our hands. Certain taxa were more common in female than in male restrooms as vagina-associated Lactobacillaceae were widely distributed in female restrooms, likely from urine contamination. Use of the SourceTracker algorithm confirmed many of our taxonomic observations as human skin was the primary source of bacteria on restroom surfaces. Overall, these results demonstrate that restroom surfaces host relatively diverse microbial communities dominated by human-associated bacteria with clear linkages between communities on or in different body sites and those communities found on restroom surfaces. More generally, this work is relevant to the public health field as we show that human associated microbes are commonly found on restroom surfaces suggesting that bacterial pathogens could readily be transmitted between individuals by the touching of surfaces. Furthermore, we demonstrate that we can use high-throughput analyses of bacterial communities to determine sources of bacteria on indoor surfaces, an approach which could be used to track pathogen transmission and test the efficacy of hygiene practices."	ENVO:building	TCAG	Microbial biogeography of public restroom surfaces	NA	0	"Ten surfaces (door handles into and out of the restroom, handles into and out of a restroom stall, faucet handles, soap dispenser, toilet seat, toilet flush handle, floor around the toilet and floor around the sink) in six male and six female restrooms evenly distributed across two buildings on the University of Colorado at Boulder campus were sampled on a single day in November 2010. Surfaces where sampled using sterile, cotton-tipped swabs as described previously?[14],?[15]. As the 12 restrooms were nearly identical in design, we were able to swab the same area at each location between restrooms. In order to characterize tap water communities as a potential source of bacteria, 1 L of faucet water from six of the restrooms (each building having the same water source for each restroom sampled) was collected and filtered through 0.2 ?m bottle top filters (Nalgene, Rochester, NY, USA)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	22132229	NA	NA	NA	Gilberto Flores	CCME	GAZ:United States of America	NA	NA	0	NA	NA	NA	NA	n	NA	Noah Fierer	Flores_restroom_surface_biogeography	pyrosequencing	NA	NA	NA	restroom surface metagenome	2	ENVO:city	Titanium	NA	NA	NA	NA	NA	flores_restroom_surface_biogeography	restroom surface biogeography	256318	CCME	NA	NA	NA	1335	"Genomic DNA was extracted from the swabs and filters using the MO BIO PowerSoil DNA isolation kit following the manufacturer's protocol with the modifications of Fierer?et al.?[14]. A portion of the 16 S rRNA gene spanning the V1_V2 regions was amplified using the primer set (27 F/338R), PCR mixture conditions and thermal cycling conditions described in Fierer?et al.?[15]. PCR amplicons of triplicate reactions for each sample were pooled at approximately equal amounts and pyrosequenced at 454 Life Sciences (Branford, CT, USA) on their GS Junior system. A total of 337,333 high-quality partial 16 S rRNA gene sequences were obtained from 101 of the 120 surface samples collected, averaging approximately 3,340 sequences per sample (ranging from 513_6,771) (Table S1) in 4 GS Junior runs, with the best run containing 116,004 high-quality reads. An additional 16,416 sequences (ranging from 2161_5084 per sample) were generated for five of the six water samples collected for source tracking analysis. Each sample was amplified with a unique barcode to enable multiplexing in the GS Junior runs. The barcoded sequencing reads can be separated by data analysis software providing high confidence in assigning sequencing read to each sample. Sequence data generated as part of this study is available upon request by contacting the corresponding author."	NA	NA	y	NA	NA	ENVO:surface	16S rRNA	Ekeley	NA	V2	CCME	NA	Toilet Floor	NA	NA	NA	40.01499	NA	NA	Floor in front of toilet
EKBM7.489474	GACTGATCATCT	YATGCTGCCTCCCGTAGGAGT	jesse.stombaugh@colorado.edu	NA	restroom surface	Floor_in_front_of_toilet	NA	NA	NA	n	12/13/10	restroom surface metagenome	NA	GFL0IUQ	NA	mimarks-survey	NA	NA	NA	NA	0	NA	NA	1624.1	10-Nov	Male	0	NA	NA	NA	NA	NA	NA	NA	NA	NA	NA	"1,swab"	B	NA	Microbial biogeography of public restroom surfaces	NA	NA	-105.27055	y	NA	CCME	NA	NA	NA	"We spend the majority of our lives indoors where we are constantly exposed to bacteria residing on surfaces. However, the diversity of these surface-associated communities is largely unknown. We explored the biogeographical patterns exhibited by bacteria across ten surfaces within each of twelve public restrooms. Using high-throughput barcoded pyrosequencing of the 16S rRNA gene, we identified 19 bacterial phyla across all surfaces. Most sequences belonged to four phyla: Actinobacteria, Bacteriodetes, Firmicutes and Proteobacteria. The communities clustered into three general categories: those found on surfaces associated with toilets, those on the restroom floor, and those found on surfaces routinely touched with hands. On toilet surfaces, gut-associated taxa were more prevalent, suggesting fecal contamination of these surfaces. Floor surfaces were the most diverse of all communities and contained several taxa commonly found in soils. Skin-associated bacteria, especially the Propionibacteriaceae, dominated surfaces routinely touched with our hands. Certain taxa were more common in female than in male restrooms as vagina-associated Lactobacillaceae were widely distributed in female restrooms, likely from urine contamination. Use of the SourceTracker algorithm confirmed many of our taxonomic observations as human skin was the primary source of bacteria on restroom surfaces. Overall, these results demonstrate that restroom surfaces host relatively diverse microbial communities dominated by human-associated bacteria with clear linkages between communities on or in different body sites and those communities found on restroom surfaces. More generally, this work is relevant to the public health field as we show that human associated microbes are commonly found on restroom surfaces suggesting that bacterial pathogens could readily be transmitted between individuals by the touching of surfaces. Furthermore, we demonstrate that we can use high-throughput analyses of bacterial communities to determine sources of bacteria on indoor surfaces, an approach which could be used to track pathogen transmission and test the efficacy of hygiene practices."	ENVO:building	TCAG	Microbial biogeography of public restroom surfaces	NA	0	"Ten surfaces (door handles into and out of the restroom, handles into and out of a restroom stall, faucet handles, soap dispenser, toilet seat, toilet flush handle, floor around the toilet and floor around the sink) in six male and six female restrooms evenly distributed across two buildings on the University of Colorado at Boulder campus were sampled on a single day in November 2010. Surfaces where sampled using sterile, cotton-tipped swabs as described previously?[14],?[15]. As the 12 restrooms were nearly identical in design, we were able to swab the same area at each location between restrooms. In order to characterize tap water communities as a potential source of bacteria, 1 L of faucet water from six of the restrooms (each building having the same water source for each restroom sampled) was collected and filtered through 0.2 ?m bottle top filters (Nalgene, Rochester, NY, USA)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	22132229	NA	NA	NA	Gilberto Flores	CCME	GAZ:United States of America	NA	NA	0	NA	NA	NA	NA	n	NA	Noah Fierer	Flores_restroom_surface_biogeography	pyrosequencing	NA	NA	NA	restroom surface metagenome	2	ENVO:city	Titanium	NA	NA	NA	NA	NA	flores_restroom_surface_biogeography	restroom surface biogeography	256318	CCME	NA	NA	NA	1335	"Genomic DNA was extracted from the swabs and filters using the MO BIO PowerSoil DNA isolation kit following the manufacturer's protocol with the modifications of Fierer?et al.?[14]. A portion of the 16 S rRNA gene spanning the V1_V2 regions was amplified using the primer set (27 F/338R), PCR mixture conditions and thermal cycling conditions described in Fierer?et al.?[15]. PCR amplicons of triplicate reactions for each sample were pooled at approximately equal amounts and pyrosequenced at 454 Life Sciences (Branford, CT, USA) on their GS Junior system. A total of 337,333 high-quality partial 16 S rRNA gene sequences were obtained from 101 of the 120 surface samples collected, averaging approximately 3,340 sequences per sample (ranging from 513_6,771) (Table S1) in 4 GS Junior runs, with the best run containing 116,004 high-quality reads. An additional 16,416 sequences (ranging from 2161_5084 per sample) were generated for five of the six water samples collected for source tracking analysis. Each sample was amplified with a unique barcode to enable multiplexing in the GS Junior runs. The barcoded sequencing reads can be separated by data analysis software providing high confidence in assigning sequencing read to each sample. Sequence data generated as part of this study is available upon request by contacting the corresponding author."	NA	NA	y	NA	NA	ENVO:surface	16S rRNA	Ekeley	NA	V2	CCME	NA	Toilet Floor	NA	NA	NA	40.01499	NA	NA	Floor in front of toilet
EKCF7.489557	GATGCATGACGC	YATGCTGCCTCCCGTAGGAGT	jesse.stombaugh@colorado.edu	NA	restroom surface	Floor_in_front_of_toilet	NA	NA	NA	n	12/13/10	restroom surface metagenome	NA	GFL0IUQ	NA	mimarks-survey	NA	NA	NA	NA	0	NA	NA	1624.1	10-Nov	Female	0	NA	NA	NA	NA	NA	NA	NA	NA	NA	NA	"1,swab"	C	NA	Microbial biogeography of public restroom surfaces	NA	NA	-105.27055	y	NA	CCME	NA	NA	NA	"We spend the majority of our lives indoors where we are constantly exposed to bacteria residing on surfaces. However, the diversity of these surface-associated communities is largely unknown. We explored the biogeographical patterns exhibited by bacteria across ten surfaces within each of twelve public restrooms. Using high-throughput barcoded pyrosequencing of the 16S rRNA gene, we identified 19 bacterial phyla across all surfaces. Most sequences belonged to four phyla: Actinobacteria, Bacteriodetes, Firmicutes and Proteobacteria. The communities clustered into three general categories: those found on surfaces associated with toilets, those on the restroom floor, and those found on surfaces routinely touched with hands. On toilet surfaces, gut-associated taxa were more prevalent, suggesting fecal contamination of these surfaces. Floor surfaces were the most diverse of all communities and contained several taxa commonly found in soils. Skin-associated bacteria, especially the Propionibacteriaceae, dominated surfaces routinely touched with our hands. Certain taxa were more common in female than in male restrooms as vagina-associated Lactobacillaceae were widely distributed in female restrooms, likely from urine contamination. Use of the SourceTracker algorithm confirmed many of our taxonomic observations as human skin was the primary source of bacteria on restroom surfaces. Overall, these results demonstrate that restroom surfaces host relatively diverse microbial communities dominated by human-associated bacteria with clear linkages between communities on or in different body sites and those communities found on restroom surfaces. More generally, this work is relevant to the public health field as we show that human associated microbes are commonly found on restroom surfaces suggesting that bacterial pathogens could readily be transmitted between individuals by the touching of surfaces. Furthermore, we demonstrate that we can use high-throughput analyses of bacterial communities to determine sources of bacteria on indoor surfaces, an approach which could be used to track pathogen transmission and test the efficacy of hygiene practices."	ENVO:building	TCAG	Microbial biogeography of public restroom surfaces	NA	0	"Ten surfaces (door handles into and out of the restroom, handles into and out of a restroom stall, faucet handles, soap dispenser, toilet seat, toilet flush handle, floor around the toilet and floor around the sink) in six male and six female restrooms evenly distributed across two buildings on the University of Colorado at Boulder campus were sampled on a single day in November 2010. Surfaces where sampled using sterile, cotton-tipped swabs as described previously?[14],?[15]. As the 12 restrooms were nearly identical in design, we were able to swab the same area at each location between restrooms. In order to characterize tap water communities as a potential source of bacteria, 1 L of faucet water from six of the restrooms (each building having the same water source for each restroom sampled) was collected and filtered through 0.2 ?m bottle top filters (Nalgene, Rochester, NY, USA)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	22132229	NA	NA	NA	Gilberto Flores	CCME	GAZ:United States of America	NA	NA	0	NA	NA	NA	NA	n	NA	Noah Fierer	Flores_restroom_surface_biogeography	pyrosequencing	NA	NA	NA	restroom surface metagenome	3	ENVO:city	Titanium	NA	NA	NA	NA	NA	flores_restroom_surface_biogeography	restroom surface biogeography	256318	CCME	NA	NA	NA	1335	"Genomic DNA was extracted from the swabs and filters using the MO BIO PowerSoil DNA isolation kit following the manufacturer's protocol with the modifications of Fierer?et al.?[14]. A portion of the 16 S rRNA gene spanning the V1_V2 regions was amplified using the primer set (27 F/338R), PCR mixture conditions and thermal cycling conditions described in Fierer?et al.?[15]. PCR amplicons of triplicate reactions for each sample were pooled at approximately equal amounts and pyrosequenced at 454 Life Sciences (Branford, CT, USA) on their GS Junior system. A total of 337,333 high-quality partial 16 S rRNA gene sequences were obtained from 101 of the 120 surface samples collected, averaging approximately 3,340 sequences per sample (ranging from 513_6,771) (Table S1) in 4 GS Junior runs, with the best run containing 116,004 high-quality reads. An additional 16,416 sequences (ranging from 2161_5084 per sample) were generated for five of the six water samples collected for source tracking analysis. Each sample was amplified with a unique barcode to enable multiplexing in the GS Junior runs. The barcoded sequencing reads can be separated by data analysis software providing high confidence in assigning sequencing read to each sample. Sequence data generated as part of this study is available upon request by contacting the corresponding author."	NA	NA	y	NA	NA	ENVO:surface	16S rRNA	Ekeley	NA	V2	CCME	NA	Toilet Floor	NA	NA	NA	40.01499	NA	NA	Floor in front of toilet
EKCM7.489464	GATAGTGCCACT	YATGCTGCCTCCCGTAGGAGT	jesse.stombaugh@colorado.edu	NA	restroom surface	Floor_in_front_of_toilet	NA	NA	NA	n	12/13/10	restroom surface metagenome	NA	GFL0IUQ	NA	mimarks-survey	NA	NA	NA	NA	0	NA	NA	1624.1	10-Nov	Male	0	NA	NA	NA	NA	NA	NA	NA	NA	NA	NA	"1,swab"	C	NA	Microbial biogeography of public restroom surfaces	NA	NA	-105.27055	y	NA	CCME	NA	NA	NA	"We spend the majority of our lives indoors where we are constantly exposed to bacteria residing on surfaces. However, the diversity of these surface-associated communities is largely unknown. We explored the biogeographical patterns exhibited by bacteria across ten surfaces within each of twelve public restrooms. Using high-throughput barcoded pyrosequencing of the 16S rRNA gene, we identified 19 bacterial phyla across all surfaces. Most sequences belonged to four phyla: Actinobacteria, Bacteriodetes, Firmicutes and Proteobacteria. The communities clustered into three general categories: those found on surfaces associated with toilets, those on the restroom floor, and those found on surfaces routinely touched with hands. On toilet surfaces, gut-associated taxa were more prevalent, suggesting fecal contamination of these surfaces. Floor surfaces were the most diverse of all communities and contained several taxa commonly found in soils. Skin-associated bacteria, especially the Propionibacteriaceae, dominated surfaces routinely touched with our hands. Certain taxa were more common in female than in male restrooms as vagina-associated Lactobacillaceae were widely distributed in female restrooms, likely from urine contamination. Use of the SourceTracker algorithm confirmed many of our taxonomic observations as human skin was the primary source of bacteria on restroom surfaces. Overall, these results demonstrate that restroom surfaces host relatively diverse microbial communities dominated by human-associated bacteria with clear linkages between communities on or in different body sites and those communities found on restroom surfaces. More generally, this work is relevant to the public health field as we show that human associated microbes are commonly found on restroom surfaces suggesting that bacterial pathogens could readily be transmitted between individuals by the touching of surfaces. Furthermore, we demonstrate that we can use high-throughput analyses of bacterial communities to determine sources of bacteria on indoor surfaces, an approach which could be used to track pathogen transmission and test the efficacy of hygiene practices."	ENVO:building	TCAG	Microbial biogeography of public restroom surfaces	NA	0	"Ten surfaces (door handles into and out of the restroom, handles into and out of a restroom stall, faucet handles, soap dispenser, toilet seat, toilet flush handle, floor around the toilet and floor around the sink) in six male and six female restrooms evenly distributed across two buildings on the University of Colorado at Boulder campus were sampled on a single day in November 2010. Surfaces where sampled using sterile, cotton-tipped swabs as described previously?[14],?[15]. As the 12 restrooms were nearly identical in design, we were able to swab the same area at each location between restrooms. In order to characterize tap water communities as a potential source of bacteria, 1 L of faucet water from six of the restrooms (each building having the same water source for each restroom sampled) was collected and filtered through 0.2 ?m bottle top filters (Nalgene, Rochester, NY, USA)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	22132229	NA	NA	NA	Gilberto Flores	CCME	GAZ:United States of America	NA	NA	0	NA	NA	NA	NA	n	NA	Noah Fierer	Flores_restroom_surface_biogeography	pyrosequencing	NA	NA	NA	restroom surface metagenome	3	ENVO:city	Titanium	NA	NA	NA	NA	NA	flores_restroom_surface_biogeography	restroom surface biogeography	256318	CCME	NA	NA	NA	1335	"Genomic DNA was extracted from the swabs and filters using the MO BIO PowerSoil DNA isolation kit following the manufacturer's protocol with the modifications of Fierer?et al.?[14]. A portion of the 16 S rRNA gene spanning the V1_V2 regions was amplified using the primer set (27 F/338R), PCR mixture conditions and thermal cycling conditions described in Fierer?et al.?[15]. PCR amplicons of triplicate reactions for each sample were pooled at approximately equal amounts and pyrosequenced at 454 Life Sciences (Branford, CT, USA) on their GS Junior system. A total of 337,333 high-quality partial 16 S rRNA gene sequences were obtained from 101 of the 120 surface samples collected, averaging approximately 3,340 sequences per sample (ranging from 513_6,771) (Table S1) in 4 GS Junior runs, with the best run containing 116,004 high-quality reads. An additional 16,416 sequences (ranging from 2161_5084 per sample) were generated for five of the six water samples collected for source tracking analysis. Each sample was amplified with a unique barcode to enable multiplexing in the GS Junior runs. The barcoded sequencing reads can be separated by data analysis software providing high confidence in assigning sequencing read to each sample. Sequence data generated as part of this study is available upon request by contacting the corresponding author."	NA	NA	y	NA	NA	ENVO:surface	16S rRNA	Ekeley	NA	V2	CCME	NA	Toilet Floor	NA	NA	NA	40.01499	NA	NA	Floor in front of toilet
PTAF7.489568	GCCACTGATAGT	YATGCTGCCTCCCGTAGGAGT	jesse.stombaugh@colorado.edu	NA	restroom surface	Floor_in_front_of_toilet	NA	NA	NA	n	12/13/10	restroom surface metagenome	NA	GFL0IUQ	NA	mimarks-survey	NA	NA	NA	NA	0	NA	NA	1624.1	10-Nov	Female	0	NA	NA	NA	NA	NA	NA	NA	NA	NA	NA	"1,swab"	A	NA	Microbial biogeography of public restroom surfaces	NA	NA	-105.27055	y	NA	CCME	NA	NA	NA	"We spend the majority of our lives indoors where we are constantly exposed to bacteria residing on surfaces. However, the diversity of these surface-associated communities is largely unknown. We explored the biogeographical patterns exhibited by bacteria across ten surfaces within each of twelve public restrooms. Using high-throughput barcoded pyrosequencing of the 16S rRNA gene, we identified 19 bacterial phyla across all surfaces. Most sequences belonged to four phyla: Actinobacteria, Bacteriodetes, Firmicutes and Proteobacteria. The communities clustered into three general categories: those found on surfaces associated with toilets, those on the restroom floor, and those found on surfaces routinely touched with hands. On toilet surfaces, gut-associated taxa were more prevalent, suggesting fecal contamination of these surfaces. Floor surfaces were the most diverse of all communities and contained several taxa commonly found in soils. Skin-associated bacteria, especially the Propionibacteriaceae, dominated surfaces routinely touched with our hands. Certain taxa were more common in female than in male restrooms as vagina-associated Lactobacillaceae were widely distributed in female restrooms, likely from urine contamination. Use of the SourceTracker algorithm confirmed many of our taxonomic observations as human skin was the primary source of bacteria on restroom surfaces. Overall, these results demonstrate that restroom surfaces host relatively diverse microbial communities dominated by human-associated bacteria with clear linkages between communities on or in different body sites and those communities found on restroom surfaces. More generally, this work is relevant to the public health field as we show that human associated microbes are commonly found on restroom surfaces suggesting that bacterial pathogens could readily be transmitted between individuals by the touching of surfaces. Furthermore, we demonstrate that we can use high-throughput analyses of bacterial communities to determine sources of bacteria on indoor surfaces, an approach which could be used to track pathogen transmission and test the efficacy of hygiene practices."	ENVO:building	TCAG	Microbial biogeography of public restroom surfaces	NA	0	"Ten surfaces (door handles into and out of the restroom, handles into and out of a restroom stall, faucet handles, soap dispenser, toilet seat, toilet flush handle, floor around the toilet and floor around the sink) in six male and six female restrooms evenly distributed across two buildings on the University of Colorado at Boulder campus were sampled on a single day in November 2010. Surfaces where sampled using sterile, cotton-tipped swabs as described previously?[14],?[15]. As the 12 restrooms were nearly identical in design, we were able to swab the same area at each location between restrooms. In order to characterize tap water communities as a potential source of bacteria, 1 L of faucet water from six of the restrooms (each building having the same water source for each restroom sampled) was collected and filtered through 0.2 ?m bottle top filters (Nalgene, Rochester, NY, USA)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	22132229	NA	NA	NA	Gilberto Flores	CCME	GAZ:United States of America	NA	NA	0	NA	NA	NA	NA	n	NA	Noah Fierer	Flores_restroom_surface_biogeography	pyrosequencing	NA	NA	NA	restroom surface metagenome	1	ENVO:city	Titanium	NA	NA	NA	NA	NA	flores_restroom_surface_biogeography	restroom surface biogeography	256318	CCME	NA	NA	NA	1335	"Genomic DNA was extracted from the swabs and filters using the MO BIO PowerSoil DNA isolation kit following the manufacturer's protocol with the modifications of Fierer?et al.?[14]. A portion of the 16 S rRNA gene spanning the V1_V2 regions was amplified using the primer set (27 F/338R), PCR mixture conditions and thermal cycling conditions described in Fierer?et al.?[15]. PCR amplicons of triplicate reactions for each sample were pooled at approximately equal amounts and pyrosequenced at 454 Life Sciences (Branford, CT, USA) on their GS Junior system. A total of 337,333 high-quality partial 16 S rRNA gene sequences were obtained from 101 of the 120 surface samples collected, averaging approximately 3,340 sequences per sample (ranging from 513_6,771) (Table S1) in 4 GS Junior runs, with the best run containing 116,004 high-quality reads. An additional 16,416 sequences (ranging from 2161_5084 per sample) were generated for five of the six water samples collected for source tracking analysis. Each sample was amplified with a unique barcode to enable multiplexing in the GS Junior runs. The barcoded sequencing reads can be separated by data analysis software providing high confidence in assigning sequencing read to each sample. Sequence data generated as part of this study is available upon request by contacting the corresponding author."	NA	NA	y	NA	NA	ENVO:surface	16S rRNA	Porter	NA	V2	CCME	NA	Toilet Floor	NA	NA	NA	40.01499	NA	NA	Floor in front of toilet
PTAM7.489553	GCAGCCGAGTAT	YATGCTGCCTCCCGTAGGAGT	jesse.stombaugh@colorado.edu	NA	restroom surface	Floor_in_front_of_toilet	NA	NA	NA	n	12/13/10	restroom surface metagenome	NA	GFL0IUQ	NA	mimarks-survey	NA	NA	NA	NA	0	NA	NA	1624.1	10-Nov	Male	0	NA	NA	NA	NA	NA	NA	NA	NA	NA	NA	"1,swab"	A	NA	Microbial biogeography of public restroom surfaces	NA	NA	-105.27055	y	NA	CCME	NA	NA	NA	"We spend the majority of our lives indoors where we are constantly exposed to bacteria residing on surfaces. However, the diversity of these surface-associated communities is largely unknown. We explored the biogeographical patterns exhibited by bacteria across ten surfaces within each of twelve public restrooms. Using high-throughput barcoded pyrosequencing of the 16S rRNA gene, we identified 19 bacterial phyla across all surfaces. Most sequences belonged to four phyla: Actinobacteria, Bacteriodetes, Firmicutes and Proteobacteria. The communities clustered into three general categories: those found on surfaces associated with toilets, those on the restroom floor, and those found on surfaces routinely touched with hands. On toilet surfaces, gut-associated taxa were more prevalent, suggesting fecal contamination of these surfaces. Floor surfaces were the most diverse of all communities and contained several taxa commonly found in soils. Skin-associated bacteria, especially the Propionibacteriaceae, dominated surfaces routinely touched with our hands. Certain taxa were more common in female than in male restrooms as vagina-associated Lactobacillaceae were widely distributed in female restrooms, likely from urine contamination. Use of the SourceTracker algorithm confirmed many of our taxonomic observations as human skin was the primary source of bacteria on restroom surfaces. Overall, these results demonstrate that restroom surfaces host relatively diverse microbial communities dominated by human-associated bacteria with clear linkages between communities on or in different body sites and those communities found on restroom surfaces. More generally, this work is relevant to the public health field as we show that human associated microbes are commonly found on restroom surfaces suggesting that bacterial pathogens could readily be transmitted between individuals by the touching of surfaces. Furthermore, we demonstrate that we can use high-throughput analyses of bacterial communities to determine sources of bacteria on indoor surfaces, an approach which could be used to track pathogen transmission and test the efficacy of hygiene practices."	ENVO:building	TCAG	Microbial biogeography of public restroom surfaces	NA	0	"Ten surfaces (door handles into and out of the restroom, handles into and out of a restroom stall, faucet handles, soap dispenser, toilet seat, toilet flush handle, floor around the toilet and floor around the sink) in six male and six female restrooms evenly distributed across two buildings on the University of Colorado at Boulder campus were sampled on a single day in November 2010. Surfaces where sampled using sterile, cotton-tipped swabs as described previously?[14],?[15]. As the 12 restrooms were nearly identical in design, we were able to swab the same area at each location between restrooms. In order to characterize tap water communities as a potential source of bacteria, 1 L of faucet water from six of the restrooms (each building having the same water source for each restroom sampled) was collected and filtered through 0.2 ?m bottle top filters (Nalgene, Rochester, NY, USA)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	22132229	NA	NA	NA	Gilberto Flores	CCME	GAZ:United States of America	NA	NA	0	NA	NA	NA	NA	n	NA	Noah Fierer	Flores_restroom_surface_biogeography	pyrosequencing	NA	NA	NA	restroom surface metagenome	1	ENVO:city	Titanium	NA	NA	NA	NA	NA	flores_restroom_surface_biogeography	restroom surface biogeography	256318	CCME	NA	NA	NA	1335	"Genomic DNA was extracted from the swabs and filters using the MO BIO PowerSoil DNA isolation kit following the manufacturer's protocol with the modifications of Fierer?et al.?[14]. A portion of the 16 S rRNA gene spanning the V1_V2 regions was amplified using the primer set (27 F/338R), PCR mixture conditions and thermal cycling conditions described in Fierer?et al.?[15]. PCR amplicons of triplicate reactions for each sample were pooled at approximately equal amounts and pyrosequenced at 454 Life Sciences (Branford, CT, USA) on their GS Junior system. A total of 337,333 high-quality partial 16 S rRNA gene sequences were obtained from 101 of the 120 surface samples collected, averaging approximately 3,340 sequences per sample (ranging from 513_6,771) (Table S1) in 4 GS Junior runs, with the best run containing 116,004 high-quality reads. An additional 16,416 sequences (ranging from 2161_5084 per sample) were generated for five of the six water samples collected for source tracking analysis. Each sample was amplified with a unique barcode to enable multiplexing in the GS Junior runs. The barcoded sequencing reads can be separated by data analysis software providing high confidence in assigning sequencing read to each sample. Sequence data generated as part of this study is available upon request by contacting the corresponding author."	NA	NA	y	NA	NA	ENVO:surface	16S rRNA	Porter	NA	V2	CCME	NA	Toilet Floor	NA	NA	NA	40.01499	NA	NA	Floor in front of toilet
PTBF7.489532	AGTGCGATGCGT	YATGCTGCCTCCCGTAGGAGT	jesse.stombaugh@colorado.edu	NA	restroom surface	Floor_in_front_of_toilet	NA	NA	NA	n	12/13/10	restroom surface metagenome	NA	GFL0IUQ	NA	mimarks-survey	NA	NA	NA	NA	0	NA	NA	1624.1	10-Nov	Female	0	NA	NA	NA	NA	NA	NA	NA	NA	NA	NA	"1,swab"	B	NA	Microbial biogeography of public restroom surfaces	NA	NA	-105.27055	y	NA	CCME	NA	NA	NA	"We spend the majority of our lives indoors where we are constantly exposed to bacteria residing on surfaces. However, the diversity of these surface-associated communities is largely unknown. We explored the biogeographical patterns exhibited by bacteria across ten surfaces within each of twelve public restrooms. Using high-throughput barcoded pyrosequencing of the 16S rRNA gene, we identified 19 bacterial phyla across all surfaces. Most sequences belonged to four phyla: Actinobacteria, Bacteriodetes, Firmicutes and Proteobacteria. The communities clustered into three general categories: those found on surfaces associated with toilets, those on the restroom floor, and those found on surfaces routinely touched with hands. On toilet surfaces, gut-associated taxa were more prevalent, suggesting fecal contamination of these surfaces. Floor surfaces were the most diverse of all communities and contained several taxa commonly found in soils. Skin-associated bacteria, especially the Propionibacteriaceae, dominated surfaces routinely touched with our hands. Certain taxa were more common in female than in male restrooms as vagina-associated Lactobacillaceae were widely distributed in female restrooms, likely from urine contamination. Use of the SourceTracker algorithm confirmed many of our taxonomic observations as human skin was the primary source of bacteria on restroom surfaces. Overall, these results demonstrate that restroom surfaces host relatively diverse microbial communities dominated by human-associated bacteria with clear linkages between communities on or in different body sites and those communities found on restroom surfaces. More generally, this work is relevant to the public health field as we show that human associated microbes are commonly found on restroom surfaces suggesting that bacterial pathogens could readily be transmitted between individuals by the touching of surfaces. Furthermore, we demonstrate that we can use high-throughput analyses of bacterial communities to determine sources of bacteria on indoor surfaces, an approach which could be used to track pathogen transmission and test the efficacy of hygiene practices."	ENVO:building	TCAG	Microbial biogeography of public restroom surfaces	NA	0	"Ten surfaces (door handles into and out of the restroom, handles into and out of a restroom stall, faucet handles, soap dispenser, toilet seat, toilet flush handle, floor around the toilet and floor around the sink) in six male and six female restrooms evenly distributed across two buildings on the University of Colorado at Boulder campus were sampled on a single day in November 2010. Surfaces where sampled using sterile, cotton-tipped swabs as described previously?[14],?[15]. As the 12 restrooms were nearly identical in design, we were able to swab the same area at each location between restrooms. In order to characterize tap water communities as a potential source of bacteria, 1 L of faucet water from six of the restrooms (each building having the same water source for each restroom sampled) was collected and filtered through 0.2 ?m bottle top filters (Nalgene, Rochester, NY, USA)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	22132229	NA	NA	NA	Gilberto Flores	CCME	GAZ:United States of America	NA	NA	0	NA	NA	NA	NA	n	NA	Noah Fierer	Flores_restroom_surface_biogeography	pyrosequencing	NA	NA	NA	restroom surface metagenome	2	ENVO:city	Titanium	NA	NA	NA	NA	NA	flores_restroom_surface_biogeography	restroom surface biogeography	256318	CCME	NA	NA	NA	1335	"Genomic DNA was extracted from the swabs and filters using the MO BIO PowerSoil DNA isolation kit following the manufacturer's protocol with the modifications of Fierer?et al.?[14]. A portion of the 16 S rRNA gene spanning the V1_V2 regions was amplified using the primer set (27 F/338R), PCR mixture conditions and thermal cycling conditions described in Fierer?et al.?[15]. PCR amplicons of triplicate reactions for each sample were pooled at approximately equal amounts and pyrosequenced at 454 Life Sciences (Branford, CT, USA) on their GS Junior system. A total of 337,333 high-quality partial 16 S rRNA gene sequences were obtained from 101 of the 120 surface samples collected, averaging approximately 3,340 sequences per sample (ranging from 513_6,771) (Table S1) in 4 GS Junior runs, with the best run containing 116,004 high-quality reads. An additional 16,416 sequences (ranging from 2161_5084 per sample) were generated for five of the six water samples collected for source tracking analysis. Each sample was amplified with a unique barcode to enable multiplexing in the GS Junior runs. The barcoded sequencing reads can be separated by data analysis software providing high confidence in assigning sequencing read to each sample. Sequence data generated as part of this study is available upon request by contacting the corresponding author."	NA	NA	y	NA	NA	ENVO:surface	16S rRNA	Porter	NA	V2	CCME	NA	Toilet Floor	NA	NA	NA	40.01499	NA	NA	Floor in front of toilet
PTBM7.489547	GCTAAGAGAGTA	YATGCTGCCTCCCGTAGGAGT	jesse.stombaugh@colorado.edu	NA	restroom surface	Floor_in_front_of_toilet	NA	NA	NA	n	12/13/10	restroom surface metagenome	NA	GFL0IUQ	NA	mimarks-survey	NA	NA	NA	NA	0	NA	NA	1624.1	10-Nov	Male	0	NA	NA	NA	NA	NA	NA	NA	NA	NA	NA	"1,swab"	B	NA	Microbial biogeography of public restroom surfaces	NA	NA	-105.27055	y	NA	CCME	NA	NA	NA	"We spend the majority of our lives indoors where we are constantly exposed to bacteria residing on surfaces. However, the diversity of these surface-associated communities is largely unknown. We explored the biogeographical patterns exhibited by bacteria across ten surfaces within each of twelve public restrooms. Using high-throughput barcoded pyrosequencing of the 16S rRNA gene, we identified 19 bacterial phyla across all surfaces. Most sequences belonged to four phyla: Actinobacteria, Bacteriodetes, Firmicutes and Proteobacteria. The communities clustered into three general categories: those found on surfaces associated with toilets, those on the restroom floor, and those found on surfaces routinely touched with hands. On toilet surfaces, gut-associated taxa were more prevalent, suggesting fecal contamination of these surfaces. Floor surfaces were the most diverse of all communities and contained several taxa commonly found in soils. Skin-associated bacteria, especially the Propionibacteriaceae, dominated surfaces routinely touched with our hands. Certain taxa were more common in female than in male restrooms as vagina-associated Lactobacillaceae were widely distributed in female restrooms, likely from urine contamination. Use of the SourceTracker algorithm confirmed many of our taxonomic observations as human skin was the primary source of bacteria on restroom surfaces. Overall, these results demonstrate that restroom surfaces host relatively diverse microbial communities dominated by human-associated bacteria with clear linkages between communities on or in different body sites and those communities found on restroom surfaces. More generally, this work is relevant to the public health field as we show that human associated microbes are commonly found on restroom surfaces suggesting that bacterial pathogens could readily be transmitted between individuals by the touching of surfaces. Furthermore, we demonstrate that we can use high-throughput analyses of bacterial communities to determine sources of bacteria on indoor surfaces, an approach which could be used to track pathogen transmission and test the efficacy of hygiene practices."	ENVO:building	TCAG	Microbial biogeography of public restroom surfaces	NA	0	"Ten surfaces (door handles into and out of the restroom, handles into and out of a restroom stall, faucet handles, soap dispenser, toilet seat, toilet flush handle, floor around the toilet and floor around the sink) in six male and six female restrooms evenly distributed across two buildings on the University of Colorado at Boulder campus were sampled on a single day in November 2010. Surfaces where sampled using sterile, cotton-tipped swabs as described previously?[14],?[15]. As the 12 restrooms were nearly identical in design, we were able to swab the same area at each location between restrooms. In order to characterize tap water communities as a potential source of bacteria, 1 L of faucet water from six of the restrooms (each building having the same water source for each restroom sampled) was collected and filtered through 0.2 ?m bottle top filters (Nalgene, Rochester, NY, USA)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	22132229	NA	NA	NA	Gilberto Flores	CCME	GAZ:United States of America	NA	NA	0	NA	NA	NA	NA	n	NA	Noah Fierer	Flores_restroom_surface_biogeography	pyrosequencing	NA	NA	NA	restroom surface metagenome	2	ENVO:city	Titanium	NA	NA	NA	NA	NA	flores_restroom_surface_biogeography	restroom surface biogeography	256318	CCME	NA	NA	NA	1335	"Genomic DNA was extracted from the swabs and filters using the MO BIO PowerSoil DNA isolation kit following the manufacturer's protocol with the modifications of Fierer?et al.?[14]. A portion of the 16 S rRNA gene spanning the V1_V2 regions was amplified using the primer set (27 F/338R), PCR mixture conditions and thermal cycling conditions described in Fierer?et al.?[15]. PCR amplicons of triplicate reactions for each sample were pooled at approximately equal amounts and pyrosequenced at 454 Life Sciences (Branford, CT, USA) on their GS Junior system. A total of 337,333 high-quality partial 16 S rRNA gene sequences were obtained from 101 of the 120 surface samples collected, averaging approximately 3,340 sequences per sample (ranging from 513_6,771) (Table S1) in 4 GS Junior runs, with the best run containing 116,004 high-quality reads. An additional 16,416 sequences (ranging from 2161_5084 per sample) were generated for five of the six water samples collected for source tracking analysis. Each sample was amplified with a unique barcode to enable multiplexing in the GS Junior runs. The barcoded sequencing reads can be separated by data analysis software providing high confidence in assigning sequencing read to each sample. Sequence data generated as part of this study is available upon request by contacting the corresponding author."	NA	NA	y	NA	NA	ENVO:surface	16S rRNA	Porter	NA	V2	CCME	NA	Toilet Floor	NA	NA	NA	40.01499	NA	NA	Floor in front of toilet
PTCF7.489531	ATCCGATCACAG	YATGCTGCCTCCCGTAGGAGT	jesse.stombaugh@colorado.edu	NA	restroom surface	Floor_in_front_of_toilet	NA	NA	NA	n	12/13/10	restroom surface metagenome	NA	GFL0IUQ	NA	mimarks-survey	NA	NA	NA	NA	0	NA	NA	1624.1	10-Nov	Female	0	NA	NA	NA	NA	NA	NA	NA	NA	NA	NA	"1,swab"	C	NA	Microbial biogeography of public restroom surfaces	NA	NA	-105.27055	y	NA	CCME	NA	NA	NA	"We spend the majority of our lives indoors where we are constantly exposed to bacteria residing on surfaces. However, the diversity of these surface-associated communities is largely unknown. We explored the biogeographical patterns exhibited by bacteria across ten surfaces within each of twelve public restrooms. Using high-throughput barcoded pyrosequencing of the 16S rRNA gene, we identified 19 bacterial phyla across all surfaces. Most sequences belonged to four phyla: Actinobacteria, Bacteriodetes, Firmicutes and Proteobacteria. The communities clustered into three general categories: those found on surfaces associated with toilets, those on the restroom floor, and those found on surfaces routinely touched with hands. On toilet surfaces, gut-associated taxa were more prevalent, suggesting fecal contamination of these surfaces. Floor surfaces were the most diverse of all communities and contained several taxa commonly found in soils. Skin-associated bacteria, especially the Propionibacteriaceae, dominated surfaces routinely touched with our hands. Certain taxa were more common in female than in male restrooms as vagina-associated Lactobacillaceae were widely distributed in female restrooms, likely from urine contamination. Use of the SourceTracker algorithm confirmed many of our taxonomic observations as human skin was the primary source of bacteria on restroom surfaces. Overall, these results demonstrate that restroom surfaces host relatively diverse microbial communities dominated by human-associated bacteria with clear linkages between communities on or in different body sites and those communities found on restroom surfaces. More generally, this work is relevant to the public health field as we show that human associated microbes are commonly found on restroom surfaces suggesting that bacterial pathogens could readily be transmitted between individuals by the touching of surfaces. Furthermore, we demonstrate that we can use high-throughput analyses of bacterial communities to determine sources of bacteria on indoor surfaces, an approach which could be used to track pathogen transmission and test the efficacy of hygiene practices."	ENVO:building	TCAG	Microbial biogeography of public restroom surfaces	NA	0	"Ten surfaces (door handles into and out of the restroom, handles into and out of a restroom stall, faucet handles, soap dispenser, toilet seat, toilet flush handle, floor around the toilet and floor around the sink) in six male and six female restrooms evenly distributed across two buildings on the University of Colorado at Boulder campus were sampled on a single day in November 2010. Surfaces where sampled using sterile, cotton-tipped swabs as described previously?[14],?[15]. As the 12 restrooms were nearly identical in design, we were able to swab the same area at each location between restrooms. In order to characterize tap water communities as a potential source of bacteria, 1 L of faucet water from six of the restrooms (each building having the same water source for each restroom sampled) was collected and filtered through 0.2 ?m bottle top filters (Nalgene, Rochester, NY, USA)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	22132229	NA	NA	NA	Gilberto Flores	CCME	GAZ:United States of America	NA	NA	0	NA	NA	NA	NA	n	NA	Noah Fierer	Flores_restroom_surface_biogeography	pyrosequencing	NA	NA	NA	restroom surface metagenome	3	ENVO:city	Titanium	NA	NA	NA	NA	NA	flores_restroom_surface_biogeography	restroom surface biogeography	256318	CCME	NA	NA	NA	1335	"Genomic DNA was extracted from the swabs and filters using the MO BIO PowerSoil DNA isolation kit following the manufacturer's protocol with the modifications of Fierer?et al.?[14]. A portion of the 16 S rRNA gene spanning the V1_V2 regions was amplified using the primer set (27 F/338R), PCR mixture conditions and thermal cycling conditions described in Fierer?et al.?[15]. PCR amplicons of triplicate reactions for each sample were pooled at approximately equal amounts and pyrosequenced at 454 Life Sciences (Branford, CT, USA) on their GS Junior system. A total of 337,333 high-quality partial 16 S rRNA gene sequences were obtained from 101 of the 120 surface samples collected, averaging approximately 3,340 sequences per sample (ranging from 513_6,771) (Table S1) in 4 GS Junior runs, with the best run containing 116,004 high-quality reads. An additional 16,416 sequences (ranging from 2161_5084 per sample) were generated for five of the six water samples collected for source tracking analysis. Each sample was amplified with a unique barcode to enable multiplexing in the GS Junior runs. The barcoded sequencing reads can be separated by data analysis software providing high confidence in assigning sequencing read to each sample. Sequence data generated as part of this study is available upon request by contacting the corresponding author."	NA	NA	y	NA	NA	ENVO:surface	16S rRNA	Porter	NA	V2	CCME	NA	Toilet Floor	NA	NA	NA	40.01499	NA	NA	Floor in front of toilet
PTCM7.489468	ATACGTCTTCGA	YATGCTGCCTCCCGTAGGAGT	jesse.stombaugh@colorado.edu	NA	restroom surface	Floor_in_front_of_toilet	NA	NA	NA	n	12/13/10	restroom surface metagenome	NA	GFL0IUQ	NA	mimarks-survey	NA	NA	NA	NA	0	NA	NA	1624.1	10-Nov	Male	0	NA	NA	NA	NA	NA	NA	NA	NA	NA	NA	"1,swab"	C	NA	Microbial biogeography of public restroom surfaces	NA	NA	-105.27055	y	NA	CCME	NA	NA	NA	"We spend the majority of our lives indoors where we are constantly exposed to bacteria residing on surfaces. However, the diversity of these surface-associated communities is largely unknown. We explored the biogeographical patterns exhibited by bacteria across ten surfaces within each of twelve public restrooms. Using high-throughput barcoded pyrosequencing of the 16S rRNA gene, we identified 19 bacterial phyla across all surfaces. Most sequences belonged to four phyla: Actinobacteria, Bacteriodetes, Firmicutes and Proteobacteria. The communities clustered into three general categories: those found on surfaces associated with toilets, those on the restroom floor, and those found on surfaces routinely touched with hands. On toilet surfaces, gut-associated taxa were more prevalent, suggesting fecal contamination of these surfaces. Floor surfaces were the most diverse of all communities and contained several taxa commonly found in soils. Skin-associated bacteria, especially the Propionibacteriaceae, dominated surfaces routinely touched with our hands. Certain taxa were more common in female than in male restrooms as vagina-associated Lactobacillaceae were widely distributed in female restrooms, likely from urine contamination. Use of the SourceTracker algorithm confirmed many of our taxonomic observations as human skin was the primary source of bacteria on restroom surfaces. Overall, these results demonstrate that restroom surfaces host relatively diverse microbial communities dominated by human-associated bacteria with clear linkages between communities on or in different body sites and those communities found on restroom surfaces. More generally, this work is relevant to the public health field as we show that human associated microbes are commonly found on restroom surfaces suggesting that bacterial pathogens could readily be transmitted between individuals by the touching of surfaces. Furthermore, we demonstrate that we can use high-throughput analyses of bacterial communities to determine sources of bacteria on indoor surfaces, an approach which could be used to track pathogen transmission and test the efficacy of hygiene practices."	ENVO:building	TCAG	Microbial biogeography of public restroom surfaces	NA	0	"Ten surfaces (door handles into and out of the restroom, handles into and out of a restroom stall, faucet handles, soap dispenser, toilet seat, toilet flush handle, floor around the toilet and floor around the sink) in six male and six female restrooms evenly distributed across two buildings on the University of Colorado at Boulder campus were sampled on a single day in November 2010. Surfaces where sampled using sterile, cotton-tipped swabs as described previously?[14],?[15]. As the 12 restrooms were nearly identical in design, we were able to swab the same area at each location between restrooms. In order to characterize tap water communities as a potential source of bacteria, 1 L of faucet water from six of the restrooms (each building having the same water source for each restroom sampled) was collected and filtered through 0.2 ?m bottle top filters (Nalgene, Rochester, NY, USA)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	22132229	NA	NA	NA	Gilberto Flores	CCME	GAZ:United States of America	NA	NA	0	NA	NA	NA	NA	n	NA	Noah Fierer	Flores_restroom_surface_biogeography	pyrosequencing	NA	NA	NA	restroom surface metagenome	3	ENVO:city	Titanium	NA	NA	NA	NA	NA	flores_restroom_surface_biogeography	restroom surface biogeography	256318	CCME	NA	NA	NA	1335	"Genomic DNA was extracted from the swabs and filters using the MO BIO PowerSoil DNA isolation kit following the manufacturer's protocol with the modifications of Fierer?et al.?[14]. A portion of the 16 S rRNA gene spanning the V1_V2 regions was amplified using the primer set (27 F/338R), PCR mixture conditions and thermal cycling conditions described in Fierer?et al.?[15]. PCR amplicons of triplicate reactions for each sample were pooled at approximately equal amounts and pyrosequenced at 454 Life Sciences (Branford, CT, USA) on their GS Junior system. A total of 337,333 high-quality partial 16 S rRNA gene sequences were obtained from 101 of the 120 surface samples collected, averaging approximately 3,340 sequences per sample (ranging from 513_6,771) (Table S1) in 4 GS Junior runs, with the best run containing 116,004 high-quality reads. An additional 16,416 sequences (ranging from 2161_5084 per sample) were generated for five of the six water samples collected for source tracking analysis. Each sample was amplified with a unique barcode to enable multiplexing in the GS Junior runs. The barcoded sequencing reads can be separated by data analysis software providing high confidence in assigning sequencing read to each sample. Sequence data generated as part of this study is available upon request by contacting the corresponding author."	NA	NA	y	NA	NA	ENVO:surface	16S rRNA	Porter	NA	V2	CCME	NA	Toilet Floor	NA	NA	NA	40.01499	NA	NA	Floor in front of toilet
EKAF9.489500	GACGATATCGCG	YATGCTGCCTCCCGTAGGAGT	jesse.stombaugh@colorado.edu	NA	restroom surface	Soap_dispenser	NA	NA	NA	n	12/13/10	restroom surface metagenome	NA	GFL0IUQ	NA	mimarks-survey	NA	NA	NA	NA	0	NA	NA	1624.1	10-Nov	Female	0	NA	NA	NA	NA	NA	NA	NA	NA	NA	NA	"1,swab"	A	NA	Microbial biogeography of public restroom surfaces	NA	NA	-105.27055	y	NA	CCME	NA	NA	NA	"We spend the majority of our lives indoors where we are constantly exposed to bacteria residing on surfaces. However, the diversity of these surface-associated communities is largely unknown. We explored the biogeographical patterns exhibited by bacteria across ten surfaces within each of twelve public restrooms. Using high-throughput barcoded pyrosequencing of the 16S rRNA gene, we identified 19 bacterial phyla across all surfaces. Most sequences belonged to four phyla: Actinobacteria, Bacteriodetes, Firmicutes and Proteobacteria. The communities clustered into three general categories: those found on surfaces associated with toilets, those on the restroom floor, and those found on surfaces routinely touched with hands. On toilet surfaces, gut-associated taxa were more prevalent, suggesting fecal contamination of these surfaces. Floor surfaces were the most diverse of all communities and contained several taxa commonly found in soils. Skin-associated bacteria, especially the Propionibacteriaceae, dominated surfaces routinely touched with our hands. Certain taxa were more common in female than in male restrooms as vagina-associated Lactobacillaceae were widely distributed in female restrooms, likely from urine contamination. Use of the SourceTracker algorithm confirmed many of our taxonomic observations as human skin was the primary source of bacteria on restroom surfaces. Overall, these results demonstrate that restroom surfaces host relatively diverse microbial communities dominated by human-associated bacteria with clear linkages between communities on or in different body sites and those communities found on restroom surfaces. More generally, this work is relevant to the public health field as we show that human associated microbes are commonly found on restroom surfaces suggesting that bacterial pathogens could readily be transmitted between individuals by the touching of surfaces. Furthermore, we demonstrate that we can use high-throughput analyses of bacterial communities to determine sources of bacteria on indoor surfaces, an approach which could be used to track pathogen transmission and test the efficacy of hygiene practices."	ENVO:building	TCAG	Microbial biogeography of public restroom surfaces	NA	0	"Ten surfaces (door handles into and out of the restroom, handles into and out of a restroom stall, faucet handles, soap dispenser, toilet seat, toilet flush handle, floor around the toilet and floor around the sink) in six male and six female restrooms evenly distributed across two buildings on the University of Colorado at Boulder campus were sampled on a single day in November 2010. Surfaces where sampled using sterile, cotton-tipped swabs as described previously?[14],?[15]. As the 12 restrooms were nearly identical in design, we were able to swab the same area at each location between restrooms. In order to characterize tap water communities as a potential source of bacteria, 1 L of faucet water from six of the restrooms (each building having the same water source for each restroom sampled) was collected and filtered through 0.2 ?m bottle top filters (Nalgene, Rochester, NY, USA)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	22132229	NA	NA	NA	Gilberto Flores	CCME	GAZ:United States of America	NA	NA	0	NA	NA	NA	NA	n	NA	Noah Fierer	Flores_restroom_surface_biogeography	pyrosequencing	NA	NA	NA	restroom surface metagenome	1	ENVO:city	Titanium	NA	NA	NA	NA	NA	flores_restroom_surface_biogeography	restroom surface biogeography	256318	CCME	NA	NA	NA	1335	"Genomic DNA was extracted from the swabs and filters using the MO BIO PowerSoil DNA isolation kit following the manufacturer's protocol with the modifications of Fierer?et al.?[14]. A portion of the 16 S rRNA gene spanning the V1_V2 regions was amplified using the primer set (27 F/338R), PCR mixture conditions and thermal cycling conditions described in Fierer?et al.?[15]. PCR amplicons of triplicate reactions for each sample were pooled at approximately equal amounts and pyrosequenced at 454 Life Sciences (Branford, CT, USA) on their GS Junior system. A total of 337,333 high-quality partial 16 S rRNA gene sequences were obtained from 101 of the 120 surface samples collected, averaging approximately 3,340 sequences per sample (ranging from 513_6,771) (Table S1) in 4 GS Junior runs, with the best run containing 116,004 high-quality reads. An additional 16,416 sequences (ranging from 2161_5084 per sample) were generated for five of the six water samples collected for source tracking analysis. Each sample was amplified with a unique barcode to enable multiplexing in the GS Junior runs. The barcoded sequencing reads can be separated by data analysis software providing high confidence in assigning sequencing read to each sample. Sequence data generated as part of this study is available upon request by contacting the corresponding author."	NA	NA	y	NA	NA	ENVO:surface	16S rRNA	Ekeley	NA	V2	CCME	NA	Soap dispenser	NA	NA	NA	40.01499	NA	NA	Soap dispenser
EKBF9.489520	GAGGCTCATCAT	YATGCTGCCTCCCGTAGGAGT	jesse.stombaugh@colorado.edu	NA	restroom surface	Soap_dispenser	NA	NA	NA	n	12/13/10	restroom surface metagenome	NA	GFL0IUQ	NA	mimarks-survey	NA	NA	NA	NA	0	NA	NA	1624.1	10-Nov	Female	0	NA	NA	NA	NA	NA	NA	NA	NA	NA	NA	"1,swab"	B	NA	Microbial biogeography of public restroom surfaces	NA	NA	-105.27055	y	NA	CCME	NA	NA	NA	"We spend the majority of our lives indoors where we are constantly exposed to bacteria residing on surfaces. However, the diversity of these surface-associated communities is largely unknown. We explored the biogeographical patterns exhibited by bacteria across ten surfaces within each of twelve public restrooms. Using high-throughput barcoded pyrosequencing of the 16S rRNA gene, we identified 19 bacterial phyla across all surfaces. Most sequences belonged to four phyla: Actinobacteria, Bacteriodetes, Firmicutes and Proteobacteria. The communities clustered into three general categories: those found on surfaces associated with toilets, those on the restroom floor, and those found on surfaces routinely touched with hands. On toilet surfaces, gut-associated taxa were more prevalent, suggesting fecal contamination of these surfaces. Floor surfaces were the most diverse of all communities and contained several taxa commonly found in soils. Skin-associated bacteria, especially the Propionibacteriaceae, dominated surfaces routinely touched with our hands. Certain taxa were more common in female than in male restrooms as vagina-associated Lactobacillaceae were widely distributed in female restrooms, likely from urine contamination. Use of the SourceTracker algorithm confirmed many of our taxonomic observations as human skin was the primary source of bacteria on restroom surfaces. Overall, these results demonstrate that restroom surfaces host relatively diverse microbial communities dominated by human-associated bacteria with clear linkages between communities on or in different body sites and those communities found on restroom surfaces. More generally, this work is relevant to the public health field as we show that human associated microbes are commonly found on restroom surfaces suggesting that bacterial pathogens could readily be transmitted between individuals by the touching of surfaces. Furthermore, we demonstrate that we can use high-throughput analyses of bacterial communities to determine sources of bacteria on indoor surfaces, an approach which could be used to track pathogen transmission and test the efficacy of hygiene practices."	ENVO:building	TCAG	Microbial biogeography of public restroom surfaces	NA	0	"Ten surfaces (door handles into and out of the restroom, handles into and out of a restroom stall, faucet handles, soap dispenser, toilet seat, toilet flush handle, floor around the toilet and floor around the sink) in six male and six female restrooms evenly distributed across two buildings on the University of Colorado at Boulder campus were sampled on a single day in November 2010. Surfaces where sampled using sterile, cotton-tipped swabs as described previously?[14],?[15]. As the 12 restrooms were nearly identical in design, we were able to swab the same area at each location between restrooms. In order to characterize tap water communities as a potential source of bacteria, 1 L of faucet water from six of the restrooms (each building having the same water source for each restroom sampled) was collected and filtered through 0.2 ?m bottle top filters (Nalgene, Rochester, NY, USA)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	22132229	NA	NA	NA	Gilberto Flores	CCME	GAZ:United States of America	NA	NA	0	NA	NA	NA	NA	n	NA	Noah Fierer	Flores_restroom_surface_biogeography	pyrosequencing	NA	NA	NA	restroom surface metagenome	2	ENVO:city	Titanium	NA	NA	NA	NA	NA	flores_restroom_surface_biogeography	restroom surface biogeography	256318	CCME	NA	NA	NA	1335	"Genomic DNA was extracted from the swabs and filters using the MO BIO PowerSoil DNA isolation kit following the manufacturer's protocol with the modifications of Fierer?et al.?[14]. A portion of the 16 S rRNA gene spanning the V1_V2 regions was amplified using the primer set (27 F/338R), PCR mixture conditions and thermal cycling conditions described in Fierer?et al.?[15]. PCR amplicons of triplicate reactions for each sample were pooled at approximately equal amounts and pyrosequenced at 454 Life Sciences (Branford, CT, USA) on their GS Junior system. A total of 337,333 high-quality partial 16 S rRNA gene sequences were obtained from 101 of the 120 surface samples collected, averaging approximately 3,340 sequences per sample (ranging from 513_6,771) (Table S1) in 4 GS Junior runs, with the best run containing 116,004 high-quality reads. An additional 16,416 sequences (ranging from 2161_5084 per sample) were generated for five of the six water samples collected for source tracking analysis. Each sample was amplified with a unique barcode to enable multiplexing in the GS Junior runs. The barcoded sequencing reads can be separated by data analysis software providing high confidence in assigning sequencing read to each sample. Sequence data generated as part of this study is available upon request by contacting the corresponding author."	NA	NA	y	NA	NA	ENVO:surface	16S rRNA	Ekeley	NA	V2	CCME	NA	Soap dispenser	NA	NA	NA	40.01499	NA	NA	Soap dispenser
EKBM9.489548	GACTGTCATGCA	YATGCTGCCTCCCGTAGGAGT	jesse.stombaugh@colorado.edu	NA	restroom surface	Soap_dispenser	NA	NA	NA	n	12/13/10	restroom surface metagenome	NA	GFL0IUQ	NA	mimarks-survey	NA	NA	NA	NA	0	NA	NA	1624.1	10-Nov	Male	0	NA	NA	NA	NA	NA	NA	NA	NA	NA	NA	"1,swab"	B	NA	Microbial biogeography of public restroom surfaces	NA	NA	-105.27055	y	NA	CCME	NA	NA	NA	"We spend the majority of our lives indoors where we are constantly exposed to bacteria residing on surfaces. However, the diversity of these surface-associated communities is largely unknown. We explored the biogeographical patterns exhibited by bacteria across ten surfaces within each of twelve public restrooms. Using high-throughput barcoded pyrosequencing of the 16S rRNA gene, we identified 19 bacterial phyla across all surfaces. Most sequences belonged to four phyla: Actinobacteria, Bacteriodetes, Firmicutes and Proteobacteria. The communities clustered into three general categories: those found on surfaces associated with toilets, those on the restroom floor, and those found on surfaces routinely touched with hands. On toilet surfaces, gut-associated taxa were more prevalent, suggesting fecal contamination of these surfaces. Floor surfaces were the most diverse of all communities and contained several taxa commonly found in soils. Skin-associated bacteria, especially the Propionibacteriaceae, dominated surfaces routinely touched with our hands. Certain taxa were more common in female than in male restrooms as vagina-associated Lactobacillaceae were widely distributed in female restrooms, likely from urine contamination. Use of the SourceTracker algorithm confirmed many of our taxonomic observations as human skin was the primary source of bacteria on restroom surfaces. Overall, these results demonstrate that restroom surfaces host relatively diverse microbial communities dominated by human-associated bacteria with clear linkages between communities on or in different body sites and those communities found on restroom surfaces. More generally, this work is relevant to the public health field as we show that human associated microbes are commonly found on restroom surfaces suggesting that bacterial pathogens could readily be transmitted between individuals by the touching of surfaces. Furthermore, we demonstrate that we can use high-throughput analyses of bacterial communities to determine sources of bacteria on indoor surfaces, an approach which could be used to track pathogen transmission and test the efficacy of hygiene practices."	ENVO:building	TCAG	Microbial biogeography of public restroom surfaces	NA	0	"Ten surfaces (door handles into and out of the restroom, handles into and out of a restroom stall, faucet handles, soap dispenser, toilet seat, toilet flush handle, floor around the toilet and floor around the sink) in six male and six female restrooms evenly distributed across two buildings on the University of Colorado at Boulder campus were sampled on a single day in November 2010. Surfaces where sampled using sterile, cotton-tipped swabs as described previously?[14],?[15]. As the 12 restrooms were nearly identical in design, we were able to swab the same area at each location between restrooms. In order to characterize tap water communities as a potential source of bacteria, 1 L of faucet water from six of the restrooms (each building having the same water source for each restroom sampled) was collected and filtered through 0.2 ?m bottle top filters (Nalgene, Rochester, NY, USA)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	22132229	NA	NA	NA	Gilberto Flores	CCME	GAZ:United States of America	NA	NA	0	NA	NA	NA	NA	n	NA	Noah Fierer	Flores_restroom_surface_biogeography	pyrosequencing	NA	NA	NA	restroom surface metagenome	2	ENVO:city	Titanium	NA	NA	NA	NA	NA	flores_restroom_surface_biogeography	restroom surface biogeography	256318	CCME	NA	NA	NA	1335	"Genomic DNA was extracted from the swabs and filters using the MO BIO PowerSoil DNA isolation kit following the manufacturer's protocol with the modifications of Fierer?et al.?[14]. A portion of the 16 S rRNA gene spanning the V1_V2 regions was amplified using the primer set (27 F/338R), PCR mixture conditions and thermal cycling conditions described in Fierer?et al.?[15]. PCR amplicons of triplicate reactions for each sample were pooled at approximately equal amounts and pyrosequenced at 454 Life Sciences (Branford, CT, USA) on their GS Junior system. A total of 337,333 high-quality partial 16 S rRNA gene sequences were obtained from 101 of the 120 surface samples collected, averaging approximately 3,340 sequences per sample (ranging from 513_6,771) (Table S1) in 4 GS Junior runs, with the best run containing 116,004 high-quality reads. An additional 16,416 sequences (ranging from 2161_5084 per sample) were generated for five of the six water samples collected for source tracking analysis. Each sample was amplified with a unique barcode to enable multiplexing in the GS Junior runs. The barcoded sequencing reads can be separated by data analysis software providing high confidence in assigning sequencing read to each sample. Sequence data generated as part of this study is available upon request by contacting the corresponding author."	NA	NA	y	NA	NA	ENVO:surface	16S rRNA	Ekeley	NA	V2	CCME	NA	Soap dispenser	NA	NA	NA	40.01499	NA	NA	Soap dispenser
EKCF9.489499	GATGTGAGCGCT	YATGCTGCCTCCCGTAGGAGT	jesse.stombaugh@colorado.edu	NA	restroom surface	Soap_dispenser	NA	NA	NA	n	12/13/10	restroom surface metagenome	NA	GFL0IUQ	NA	mimarks-survey	NA	NA	NA	NA	0	NA	NA	1624.1	10-Nov	Female	0	NA	NA	NA	NA	NA	NA	NA	NA	NA	NA	"1,swab"	C	NA	Microbial biogeography of public restroom surfaces	NA	NA	-105.27055	y	NA	CCME	NA	NA	NA	"We spend the majority of our lives indoors where we are constantly exposed to bacteria residing on surfaces. However, the diversity of these surface-associated communities is largely unknown. We explored the biogeographical patterns exhibited by bacteria across ten surfaces within each of twelve public restrooms. Using high-throughput barcoded pyrosequencing of the 16S rRNA gene, we identified 19 bacterial phyla across all surfaces. Most sequences belonged to four phyla: Actinobacteria, Bacteriodetes, Firmicutes and Proteobacteria. The communities clustered into three general categories: those found on surfaces associated with toilets, those on the restroom floor, and those found on surfaces routinely touched with hands. On toilet surfaces, gut-associated taxa were more prevalent, suggesting fecal contamination of these surfaces. Floor surfaces were the most diverse of all communities and contained several taxa commonly found in soils. Skin-associated bacteria, especially the Propionibacteriaceae, dominated surfaces routinely touched with our hands. Certain taxa were more common in female than in male restrooms as vagina-associated Lactobacillaceae were widely distributed in female restrooms, likely from urine contamination. Use of the SourceTracker algorithm confirmed many of our taxonomic observations as human skin was the primary source of bacteria on restroom surfaces. Overall, these results demonstrate that restroom surfaces host relatively diverse microbial communities dominated by human-associated bacteria with clear linkages between communities on or in different body sites and those communities found on restroom surfaces. More generally, this work is relevant to the public health field as we show that human associated microbes are commonly found on restroom surfaces suggesting that bacterial pathogens could readily be transmitted between individuals by the touching of surfaces. Furthermore, we demonstrate that we can use high-throughput analyses of bacterial communities to determine sources of bacteria on indoor surfaces, an approach which could be used to track pathogen transmission and test the efficacy of hygiene practices."	ENVO:building	TCAG	Microbial biogeography of public restroom surfaces	NA	0	"Ten surfaces (door handles into and out of the restroom, handles into and out of a restroom stall, faucet handles, soap dispenser, toilet seat, toilet flush handle, floor around the toilet and floor around the sink) in six male and six female restrooms evenly distributed across two buildings on the University of Colorado at Boulder campus were sampled on a single day in November 2010. Surfaces where sampled using sterile, cotton-tipped swabs as described previously?[14],?[15]. As the 12 restrooms were nearly identical in design, we were able to swab the same area at each location between restrooms. In order to characterize tap water communities as a potential source of bacteria, 1 L of faucet water from six of the restrooms (each building having the same water source for each restroom sampled) was collected and filtered through 0.2 ?m bottle top filters (Nalgene, Rochester, NY, USA)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	22132229	NA	NA	NA	Gilberto Flores	CCME	GAZ:United States of America	NA	NA	0	NA	NA	NA	NA	n	NA	Noah Fierer	Flores_restroom_surface_biogeography	pyrosequencing	NA	NA	NA	restroom surface metagenome	3	ENVO:city	Titanium	NA	NA	NA	NA	NA	flores_restroom_surface_biogeography	restroom surface biogeography	256318	CCME	NA	NA	NA	1335	"Genomic DNA was extracted from the swabs and filters using the MO BIO PowerSoil DNA isolation kit following the manufacturer's protocol with the modifications of Fierer?et al.?[14]. A portion of the 16 S rRNA gene spanning the V1_V2 regions was amplified using the primer set (27 F/338R), PCR mixture conditions and thermal cycling conditions described in Fierer?et al.?[15]. PCR amplicons of triplicate reactions for each sample were pooled at approximately equal amounts and pyrosequenced at 454 Life Sciences (Branford, CT, USA) on their GS Junior system. A total of 337,333 high-quality partial 16 S rRNA gene sequences were obtained from 101 of the 120 surface samples collected, averaging approximately 3,340 sequences per sample (ranging from 513_6,771) (Table S1) in 4 GS Junior runs, with the best run containing 116,004 high-quality reads. An additional 16,416 sequences (ranging from 2161_5084 per sample) were generated for five of the six water samples collected for source tracking analysis. Each sample was amplified with a unique barcode to enable multiplexing in the GS Junior runs. The barcoded sequencing reads can be separated by data analysis software providing high confidence in assigning sequencing read to each sample. Sequence data generated as part of this study is available upon request by contacting the corresponding author."	NA	NA	y	NA	NA	ENVO:surface	16S rRNA	Ekeley	NA	V2	CCME	NA	Soap dispenser	NA	NA	NA	40.01499	NA	NA	Soap dispenser
PTAF9.489513	GCCTATACTACA	YATGCTGCCTCCCGTAGGAGT	jesse.stombaugh@colorado.edu	NA	restroom surface	Soap_dispenser	NA	NA	NA	n	12/13/10	restroom surface metagenome	NA	GFL0IUQ	NA	mimarks-survey	NA	NA	NA	NA	0	NA	NA	1624.1	10-Nov	Female	0	NA	NA	NA	NA	NA	NA	NA	NA	NA	NA	"1,swab"	A	NA	Microbial biogeography of public restroom surfaces	NA	NA	-105.27055	y	NA	CCME	NA	NA	NA	"We spend the majority of our lives indoors where we are constantly exposed to bacteria residing on surfaces. However, the diversity of these surface-associated communities is largely unknown. We explored the biogeographical patterns exhibited by bacteria across ten surfaces within each of twelve public restrooms. Using high-throughput barcoded pyrosequencing of the 16S rRNA gene, we identified 19 bacterial phyla across all surfaces. Most sequences belonged to four phyla: Actinobacteria, Bacteriodetes, Firmicutes and Proteobacteria. The communities clustered into three general categories: those found on surfaces associated with toilets, those on the restroom floor, and those found on surfaces routinely touched with hands. On toilet surfaces, gut-associated taxa were more prevalent, suggesting fecal contamination of these surfaces. Floor surfaces were the most diverse of all communities and contained several taxa commonly found in soils. Skin-associated bacteria, especially the Propionibacteriaceae, dominated surfaces routinely touched with our hands. Certain taxa were more common in female than in male restrooms as vagina-associated Lactobacillaceae were widely distributed in female restrooms, likely from urine contamination. Use of the SourceTracker algorithm confirmed many of our taxonomic observations as human skin was the primary source of bacteria on restroom surfaces. Overall, these results demonstrate that restroom surfaces host relatively diverse microbial communities dominated by human-associated bacteria with clear linkages between communities on or in different body sites and those communities found on restroom surfaces. More generally, this work is relevant to the public health field as we show that human associated microbes are commonly found on restroom surfaces suggesting that bacterial pathogens could readily be transmitted between individuals by the touching of surfaces. Furthermore, we demonstrate that we can use high-throughput analyses of bacterial communities to determine sources of bacteria on indoor surfaces, an approach which could be used to track pathogen transmission and test the efficacy of hygiene practices."	ENVO:building	TCAG	Microbial biogeography of public restroom surfaces	NA	0	"Ten surfaces (door handles into and out of the restroom, handles into and out of a restroom stall, faucet handles, soap dispenser, toilet seat, toilet flush handle, floor around the toilet and floor around the sink) in six male and six female restrooms evenly distributed across two buildings on the University of Colorado at Boulder campus were sampled on a single day in November 2010. Surfaces where sampled using sterile, cotton-tipped swabs as described previously?[14],?[15]. As the 12 restrooms were nearly identical in design, we were able to swab the same area at each location between restrooms. In order to characterize tap water communities as a potential source of bacteria, 1 L of faucet water from six of the restrooms (each building having the same water source for each restroom sampled) was collected and filtered through 0.2 ?m bottle top filters (Nalgene, Rochester, NY, USA)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	22132229	NA	NA	NA	Gilberto Flores	CCME	GAZ:United States of America	NA	NA	0	NA	NA	NA	NA	n	NA	Noah Fierer	Flores_restroom_surface_biogeography	pyrosequencing	NA	NA	NA	restroom surface metagenome	1	ENVO:city	Titanium	NA	NA	NA	NA	NA	flores_restroom_surface_biogeography	restroom surface biogeography	256318	CCME	NA	NA	NA	1335	"Genomic DNA was extracted from the swabs and filters using the MO BIO PowerSoil DNA isolation kit following the manufacturer's protocol with the modifications of Fierer?et al.?[14]. A portion of the 16 S rRNA gene spanning the V1_V2 regions was amplified using the primer set (27 F/338R), PCR mixture conditions and thermal cycling conditions described in Fierer?et al.?[15]. PCR amplicons of triplicate reactions for each sample were pooled at approximately equal amounts and pyrosequenced at 454 Life Sciences (Branford, CT, USA) on their GS Junior system. A total of 337,333 high-quality partial 16 S rRNA gene sequences were obtained from 101 of the 120 surface samples collected, averaging approximately 3,340 sequences per sample (ranging from 513_6,771) (Table S1) in 4 GS Junior runs, with the best run containing 116,004 high-quality reads. An additional 16,416 sequences (ranging from 2161_5084 per sample) were generated for five of the six water samples collected for source tracking analysis. Each sample was amplified with a unique barcode to enable multiplexing in the GS Junior runs. The barcoded sequencing reads can be separated by data analysis software providing high confidence in assigning sequencing read to each sample. Sequence data generated as part of this study is available upon request by contacting the corresponding author."	NA	NA	y	NA	NA	ENVO:surface	16S rRNA	Porter	NA	V2	CCME	NA	Soap dispenser	NA	NA	NA	40.01499	NA	NA	Soap dispenser
PTAM9.489543	GCAGGCAGTACT	YATGCTGCCTCCCGTAGGAGT	jesse.stombaugh@colorado.edu	NA	restroom surface	Soap_dispenser	NA	NA	NA	n	12/13/10	restroom surface metagenome	NA	GFL0IUQ	NA	mimarks-survey	NA	NA	NA	NA	0	NA	NA	1624.1	10-Nov	Male	0	NA	NA	NA	NA	NA	NA	NA	NA	NA	NA	"1,swab"	A	NA	Microbial biogeography of public restroom surfaces	NA	NA	-105.27055	y	NA	CCME	NA	NA	NA	"We spend the majority of our lives indoors where we are constantly exposed to bacteria residing on surfaces. However, the diversity of these surface-associated communities is largely unknown. We explored the biogeographical patterns exhibited by bacteria across ten surfaces within each of twelve public restrooms. Using high-throughput barcoded pyrosequencing of the 16S rRNA gene, we identified 19 bacterial phyla across all surfaces. Most sequences belonged to four phyla: Actinobacteria, Bacteriodetes, Firmicutes and Proteobacteria. The communities clustered into three general categories: those found on surfaces associated with toilets, those on the restroom floor, and those found on surfaces routinely touched with hands. On toilet surfaces, gut-associated taxa were more prevalent, suggesting fecal contamination of these surfaces. Floor surfaces were the most diverse of all communities and contained several taxa commonly found in soils. Skin-associated bacteria, especially the Propionibacteriaceae, dominated surfaces routinely touched with our hands. Certain taxa were more common in female than in male restrooms as vagina-associated Lactobacillaceae were widely distributed in female restrooms, likely from urine contamination. Use of the SourceTracker algorithm confirmed many of our taxonomic observations as human skin was the primary source of bacteria on restroom surfaces. Overall, these results demonstrate that restroom surfaces host relatively diverse microbial communities dominated by human-associated bacteria with clear linkages between communities on or in different body sites and those communities found on restroom surfaces. More generally, this work is relevant to the public health field as we show that human associated microbes are commonly found on restroom surfaces suggesting that bacterial pathogens could readily be transmitted between individuals by the touching of surfaces. Furthermore, we demonstrate that we can use high-throughput analyses of bacterial communities to determine sources of bacteria on indoor surfaces, an approach which could be used to track pathogen transmission and test the efficacy of hygiene practices."	ENVO:building	TCAG	Microbial biogeography of public restroom surfaces	NA	0	"Ten surfaces (door handles into and out of the restroom, handles into and out of a restroom stall, faucet handles, soap dispenser, toilet seat, toilet flush handle, floor around the toilet and floor around the sink) in six male and six female restrooms evenly distributed across two buildings on the University of Colorado at Boulder campus were sampled on a single day in November 2010. Surfaces where sampled using sterile, cotton-tipped swabs as described previously?[14],?[15]. As the 12 restrooms were nearly identical in design, we were able to swab the same area at each location between restrooms. In order to characterize tap water communities as a potential source of bacteria, 1 L of faucet water from six of the restrooms (each building having the same water source for each restroom sampled) was collected and filtered through 0.2 ?m bottle top filters (Nalgene, Rochester, NY, USA)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	22132229	NA	NA	NA	Gilberto Flores	CCME	GAZ:United States of America	NA	NA	0	NA	NA	NA	NA	n	NA	Noah Fierer	Flores_restroom_surface_biogeography	pyrosequencing	NA	NA	NA	restroom surface metagenome	1	ENVO:city	Titanium	NA	NA	NA	NA	NA	flores_restroom_surface_biogeography	restroom surface biogeography	256318	CCME	NA	NA	NA	1335	"Genomic DNA was extracted from the swabs and filters using the MO BIO PowerSoil DNA isolation kit following the manufacturer's protocol with the modifications of Fierer?et al.?[14]. A portion of the 16 S rRNA gene spanning the V1_V2 regions was amplified using the primer set (27 F/338R), PCR mixture conditions and thermal cycling conditions described in Fierer?et al.?[15]. PCR amplicons of triplicate reactions for each sample were pooled at approximately equal amounts and pyrosequenced at 454 Life Sciences (Branford, CT, USA) on their GS Junior system. A total of 337,333 high-quality partial 16 S rRNA gene sequences were obtained from 101 of the 120 surface samples collected, averaging approximately 3,340 sequences per sample (ranging from 513_6,771) (Table S1) in 4 GS Junior runs, with the best run containing 116,004 high-quality reads. An additional 16,416 sequences (ranging from 2161_5084 per sample) were generated for five of the six water samples collected for source tracking analysis. Each sample was amplified with a unique barcode to enable multiplexing in the GS Junior runs. The barcoded sequencing reads can be separated by data analysis software providing high confidence in assigning sequencing read to each sample. Sequence data generated as part of this study is available upon request by contacting the corresponding author."	NA	NA	y	NA	NA	ENVO:surface	16S rRNA	Porter	NA	V2	CCME	NA	Soap dispenser	NA	NA	NA	40.01499	NA	NA	Soap dispenser
PTBF9.489570	AGTGTCACGGTG	YATGCTGCCTCCCGTAGGAGT	jesse.stombaugh@colorado.edu	NA	restroom surface	Soap_dispenser	NA	NA	NA	n	12/13/10	restroom surface metagenome	NA	GFL0IUQ	NA	mimarks-survey	NA	NA	NA	NA	0	NA	NA	1624.1	10-Nov	Female	0	NA	NA	NA	NA	NA	NA	NA	NA	NA	NA	"1,swab"	B	NA	Microbial biogeography of public restroom surfaces	NA	NA	-105.27055	y	NA	CCME	NA	NA	NA	"We spend the majority of our lives indoors where we are constantly exposed to bacteria residing on surfaces. However, the diversity of these surface-associated communities is largely unknown. We explored the biogeographical patterns exhibited by bacteria across ten surfaces within each of twelve public restrooms. Using high-throughput barcoded pyrosequencing of the 16S rRNA gene, we identified 19 bacterial phyla across all surfaces. Most sequences belonged to four phyla: Actinobacteria, Bacteriodetes, Firmicutes and Proteobacteria. The communities clustered into three general categories: those found on surfaces associated with toilets, those on the restroom floor, and those found on surfaces routinely touched with hands. On toilet surfaces, gut-associated taxa were more prevalent, suggesting fecal contamination of these surfaces. Floor surfaces were the most diverse of all communities and contained several taxa commonly found in soils. Skin-associated bacteria, especially the Propionibacteriaceae, dominated surfaces routinely touched with our hands. Certain taxa were more common in female than in male restrooms as vagina-associated Lactobacillaceae were widely distributed in female restrooms, likely from urine contamination. Use of the SourceTracker algorithm confirmed many of our taxonomic observations as human skin was the primary source of bacteria on restroom surfaces. Overall, these results demonstrate that restroom surfaces host relatively diverse microbial communities dominated by human-associated bacteria with clear linkages between communities on or in different body sites and those communities found on restroom surfaces. More generally, this work is relevant to the public health field as we show that human associated microbes are commonly found on restroom surfaces suggesting that bacterial pathogens could readily be transmitted between individuals by the touching of surfaces. Furthermore, we demonstrate that we can use high-throughput analyses of bacterial communities to determine sources of bacteria on indoor surfaces, an approach which could be used to track pathogen transmission and test the efficacy of hygiene practices."	ENVO:building	TCAG	Microbial biogeography of public restroom surfaces	NA	0	"Ten surfaces (door handles into and out of the restroom, handles into and out of a restroom stall, faucet handles, soap dispenser, toilet seat, toilet flush handle, floor around the toilet and floor around the sink) in six male and six female restrooms evenly distributed across two buildings on the University of Colorado at Boulder campus were sampled on a single day in November 2010. Surfaces where sampled using sterile, cotton-tipped swabs as described previously?[14],?[15]. As the 12 restrooms were nearly identical in design, we were able to swab the same area at each location between restrooms. In order to characterize tap water communities as a potential source of bacteria, 1 L of faucet water from six of the restrooms (each building having the same water source for each restroom sampled) was collected and filtered through 0.2 ?m bottle top filters (Nalgene, Rochester, NY, USA)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	22132229	NA	NA	NA	Gilberto Flores	CCME	GAZ:United States of America	NA	NA	0	NA	NA	NA	NA	n	NA	Noah Fierer	Flores_restroom_surface_biogeography	pyrosequencing	NA	NA	NA	restroom surface metagenome	2	ENVO:city	Titanium	NA	NA	NA	NA	NA	flores_restroom_surface_biogeography	restroom surface biogeography	256318	CCME	NA	NA	NA	1335	"Genomic DNA was extracted from the swabs and filters using the MO BIO PowerSoil DNA isolation kit following the manufacturer's protocol with the modifications of Fierer?et al.?[14]. A portion of the 16 S rRNA gene spanning the V1_V2 regions was amplified using the primer set (27 F/338R), PCR mixture conditions and thermal cycling conditions described in Fierer?et al.?[15]. PCR amplicons of triplicate reactions for each sample were pooled at approximately equal amounts and pyrosequenced at 454 Life Sciences (Branford, CT, USA) on their GS Junior system. A total of 337,333 high-quality partial 16 S rRNA gene sequences were obtained from 101 of the 120 surface samples collected, averaging approximately 3,340 sequences per sample (ranging from 513_6,771) (Table S1) in 4 GS Junior runs, with the best run containing 116,004 high-quality reads. An additional 16,416 sequences (ranging from 2161_5084 per sample) were generated for five of the six water samples collected for source tracking analysis. Each sample was amplified with a unique barcode to enable multiplexing in the GS Junior runs. The barcoded sequencing reads can be separated by data analysis software providing high confidence in assigning sequencing read to each sample. Sequence data generated as part of this study is available upon request by contacting the corresponding author."	NA	NA	y	NA	NA	ENVO:surface	16S rRNA	Porter	NA	V2	CCME	NA	Soap dispenser	NA	NA	NA	40.01499	NA	NA	Soap dispenser
PTBM9.489505	GCTAGTCTGAAC	YATGCTGCCTCCCGTAGGAGT	jesse.stombaugh@colorado.edu	NA	restroom surface	Soap_dispenser	NA	NA	NA	n	12/13/10	restroom surface metagenome	NA	GFL0IUQ	NA	mimarks-survey	NA	NA	NA	NA	0	NA	NA	1624.1	10-Nov	Male	0	NA	NA	NA	NA	NA	NA	NA	NA	NA	NA	"1,swab"	B	NA	Microbial biogeography of public restroom surfaces	NA	NA	-105.27055	y	NA	CCME	NA	NA	NA	"We spend the majority of our lives indoors where we are constantly exposed to bacteria residing on surfaces. However, the diversity of these surface-associated communities is largely unknown. We explored the biogeographical patterns exhibited by bacteria across ten surfaces within each of twelve public restrooms. Using high-throughput barcoded pyrosequencing of the 16S rRNA gene, we identified 19 bacterial phyla across all surfaces. Most sequences belonged to four phyla: Actinobacteria, Bacteriodetes, Firmicutes and Proteobacteria. The communities clustered into three general categories: those found on surfaces associated with toilets, those on the restroom floor, and those found on surfaces routinely touched with hands. On toilet surfaces, gut-associated taxa were more prevalent, suggesting fecal contamination of these surfaces. Floor surfaces were the most diverse of all communities and contained several taxa commonly found in soils. Skin-associated bacteria, especially the Propionibacteriaceae, dominated surfaces routinely touched with our hands. Certain taxa were more common in female than in male restrooms as vagina-associated Lactobacillaceae were widely distributed in female restrooms, likely from urine contamination. Use of the SourceTracker algorithm confirmed many of our taxonomic observations as human skin was the primary source of bacteria on restroom surfaces. Overall, these results demonstrate that restroom surfaces host relatively diverse microbial communities dominated by human-associated bacteria with clear linkages between communities on or in different body sites and those communities found on restroom surfaces. More generally, this work is relevant to the public health field as we show that human associated microbes are commonly found on restroom surfaces suggesting that bacterial pathogens could readily be transmitted between individuals by the touching of surfaces. Furthermore, we demonstrate that we can use high-throughput analyses of bacterial communities to determine sources of bacteria on indoor surfaces, an approach which could be used to track pathogen transmission and test the efficacy of hygiene practices."	ENVO:building	TCAG	Microbial biogeography of public restroom surfaces	NA	0	"Ten surfaces (door handles into and out of the restroom, handles into and out of a restroom stall, faucet handles, soap dispenser, toilet seat, toilet flush handle, floor around the toilet and floor around the sink) in six male and six female restrooms evenly distributed across two buildings on the University of Colorado at Boulder campus were sampled on a single day in November 2010. Surfaces where sampled using sterile, cotton-tipped swabs as described previously?[14],?[15]. As the 12 restrooms were nearly identical in design, we were able to swab the same area at each location between restrooms. In order to characterize tap water communities as a potential source of bacteria, 1 L of faucet water from six of the restrooms (each building having the same water source for each restroom sampled) was collected and filtered through 0.2 ?m bottle top filters (Nalgene, Rochester, NY, USA)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	22132229	NA	NA	NA	Gilberto Flores	CCME	GAZ:United States of America	NA	NA	0	NA	NA	NA	NA	n	NA	Noah Fierer	Flores_restroom_surface_biogeography	pyrosequencing	NA	NA	NA	restroom surface metagenome	2	ENVO:city	Titanium	NA	NA	NA	NA	NA	flores_restroom_surface_biogeography	restroom surface biogeography	256318	CCME	NA	NA	NA	1335	"Genomic DNA was extracted from the swabs and filters using the MO BIO PowerSoil DNA isolation kit following the manufacturer's protocol with the modifications of Fierer?et al.?[14]. A portion of the 16 S rRNA gene spanning the V1_V2 regions was amplified using the primer set (27 F/338R), PCR mixture conditions and thermal cycling conditions described in Fierer?et al.?[15]. PCR amplicons of triplicate reactions for each sample were pooled at approximately equal amounts and pyrosequenced at 454 Life Sciences (Branford, CT, USA) on their GS Junior system. A total of 337,333 high-quality partial 16 S rRNA gene sequences were obtained from 101 of the 120 surface samples collected, averaging approximately 3,340 sequences per sample (ranging from 513_6,771) (Table S1) in 4 GS Junior runs, with the best run containing 116,004 high-quality reads. An additional 16,416 sequences (ranging from 2161_5084 per sample) were generated for five of the six water samples collected for source tracking analysis. Each sample was amplified with a unique barcode to enable multiplexing in the GS Junior runs. The barcoded sequencing reads can be separated by data analysis software providing high confidence in assigning sequencing read to each sample. Sequence data generated as part of this study is available upon request by contacting the corresponding author."	NA	NA	y	NA	NA	ENVO:surface	16S rRNA	Porter	NA	V2	CCME	NA	Soap dispenser	NA	NA	NA	40.01499	NA	NA	Soap dispenser
PTCF9.489454	ATCGATCTGTGG	YATGCTGCCTCCCGTAGGAGT	jesse.stombaugh@colorado.edu	NA	restroom surface	Soap_dispenser	NA	NA	NA	n	12/13/10	restroom surface metagenome	NA	GFL0IUQ	NA	mimarks-survey	NA	NA	NA	NA	0	NA	NA	1624.1	10-Nov	Female	0	NA	NA	NA	NA	NA	NA	NA	NA	NA	NA	"1,swab"	C	NA	Microbial biogeography of public restroom surfaces	NA	NA	-105.27055	y	NA	CCME	NA	NA	NA	"We spend the majority of our lives indoors where we are constantly exposed to bacteria residing on surfaces. However, the diversity of these surface-associated communities is largely unknown. We explored the biogeographical patterns exhibited by bacteria across ten surfaces within each of twelve public restrooms. Using high-throughput barcoded pyrosequencing of the 16S rRNA gene, we identified 19 bacterial phyla across all surfaces. Most sequences belonged to four phyla: Actinobacteria, Bacteriodetes, Firmicutes and Proteobacteria. The communities clustered into three general categories: those found on surfaces associated with toilets, those on the restroom floor, and those found on surfaces routinely touched with hands. On toilet surfaces, gut-associated taxa were more prevalent, suggesting fecal contamination of these surfaces. Floor surfaces were the most diverse of all communities and contained several taxa commonly found in soils. Skin-associated bacteria, especially the Propionibacteriaceae, dominated surfaces routinely touched with our hands. Certain taxa were more common in female than in male restrooms as vagina-associated Lactobacillaceae were widely distributed in female restrooms, likely from urine contamination. Use of the SourceTracker algorithm confirmed many of our taxonomic observations as human skin was the primary source of bacteria on restroom surfaces. Overall, these results demonstrate that restroom surfaces host relatively diverse microbial communities dominated by human-associated bacteria with clear linkages between communities on or in different body sites and those communities found on restroom surfaces. More generally, this work is relevant to the public health field as we show that human associated microbes are commonly found on restroom surfaces suggesting that bacterial pathogens could readily be transmitted between individuals by the touching of surfaces. Furthermore, we demonstrate that we can use high-throughput analyses of bacterial communities to determine sources of bacteria on indoor surfaces, an approach which could be used to track pathogen transmission and test the efficacy of hygiene practices."	ENVO:building	TCAG	Microbial biogeography of public restroom surfaces	NA	0	"Ten surfaces (door handles into and out of the restroom, handles into and out of a restroom stall, faucet handles, soap dispenser, toilet seat, toilet flush handle, floor around the toilet and floor around the sink) in six male and six female restrooms evenly distributed across two buildings on the University of Colorado at Boulder campus were sampled on a single day in November 2010. Surfaces where sampled using sterile, cotton-tipped swabs as described previously?[14],?[15]. As the 12 restrooms were nearly identical in design, we were able to swab the same area at each location between restrooms. In order to characterize tap water communities as a potential source of bacteria, 1 L of faucet water from six of the restrooms (each building having the same water source for each restroom sampled) was collected and filtered through 0.2 ?m bottle top filters (Nalgene, Rochester, NY, USA)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	22132229	NA	NA	NA	Gilberto Flores	CCME	GAZ:United States of America	NA	NA	0	NA	NA	NA	NA	n	NA	Noah Fierer	Flores_restroom_surface_biogeography	pyrosequencing	NA	NA	NA	restroom surface metagenome	3	ENVO:city	Titanium	NA	NA	NA	NA	NA	flores_restroom_surface_biogeography	restroom surface biogeography	256318	CCME	NA	NA	NA	1335	"Genomic DNA was extracted from the swabs and filters using the MO BIO PowerSoil DNA isolation kit following the manufacturer's protocol with the modifications of Fierer?et al.?[14]. A portion of the 16 S rRNA gene spanning the V1_V2 regions was amplified using the primer set (27 F/338R), PCR mixture conditions and thermal cycling conditions described in Fierer?et al.?[15]. PCR amplicons of triplicate reactions for each sample were pooled at approximately equal amounts and pyrosequenced at 454 Life Sciences (Branford, CT, USA) on their GS Junior system. A total of 337,333 high-quality partial 16 S rRNA gene sequences were obtained from 101 of the 120 surface samples collected, averaging approximately 3,340 sequences per sample (ranging from 513_6,771) (Table S1) in 4 GS Junior runs, with the best run containing 116,004 high-quality reads. An additional 16,416 sequences (ranging from 2161_5084 per sample) were generated for five of the six water samples collected for source tracking analysis. Each sample was amplified with a unique barcode to enable multiplexing in the GS Junior runs. The barcoded sequencing reads can be separated by data analysis software providing high confidence in assigning sequencing read to each sample. Sequence data generated as part of this study is available upon request by contacting the corresponding author."	NA	NA	y	NA	NA	ENVO:surface	16S rRNA	Porter	NA	V2	CCME	NA	Soap dispenser	NA	NA	NA	40.01499	NA	NA	Soap dispenser
PTCM9.489527	ATACTCACTCAG	YATGCTGCCTCCCGTAGGAGT	jesse.stombaugh@colorado.edu	NA	restroom surface	Soap_dispenser	NA	NA	NA	n	12/13/10	restroom surface metagenome	NA	GFL0IUQ	NA	mimarks-survey	NA	NA	NA	NA	0	NA	NA	1624.1	10-Nov	Male	0	NA	NA	NA	NA	NA	NA	NA	NA	NA	NA	"1,swab"	C	NA	Microbial biogeography of public restroom surfaces	NA	NA	-105.27055	y	NA	CCME	NA	NA	NA	"We spend the majority of our lives indoors where we are constantly exposed to bacteria residing on surfaces. However, the diversity of these surface-associated communities is largely unknown. We explored the biogeographical patterns exhibited by bacteria across ten surfaces within each of twelve public restrooms. Using high-throughput barcoded pyrosequencing of the 16S rRNA gene, we identified 19 bacterial phyla across all surfaces. Most sequences belonged to four phyla: Actinobacteria, Bacteriodetes, Firmicutes and Proteobacteria. The communities clustered into three general categories: those found on surfaces associated with toilets, those on the restroom floor, and those found on surfaces routinely touched with hands. On toilet surfaces, gut-associated taxa were more prevalent, suggesting fecal contamination of these surfaces. Floor surfaces were the most diverse of all communities and contained several taxa commonly found in soils. Skin-associated bacteria, especially the Propionibacteriaceae, dominated surfaces routinely touched with our hands. Certain taxa were more common in female than in male restrooms as vagina-associated Lactobacillaceae were widely distributed in female restrooms, likely from urine contamination. Use of the SourceTracker algorithm confirmed many of our taxonomic observations as human skin was the primary source of bacteria on restroom surfaces. Overall, these results demonstrate that restroom surfaces host relatively diverse microbial communities dominated by human-associated bacteria with clear linkages between communities on or in different body sites and those communities found on restroom surfaces. More generally, this work is relevant to the public health field as we show that human associated microbes are commonly found on restroom surfaces suggesting that bacterial pathogens could readily be transmitted between individuals by the touching of surfaces. Furthermore, we demonstrate that we can use high-throughput analyses of bacterial communities to determine sources of bacteria on indoor surfaces, an approach which could be used to track pathogen transmission and test the efficacy of hygiene practices."	ENVO:building	TCAG	Microbial biogeography of public restroom surfaces	NA	0	"Ten surfaces (door handles into and out of the restroom, handles into and out of a restroom stall, faucet handles, soap dispenser, toilet seat, toilet flush handle, floor around the toilet and floor around the sink) in six male and six female restrooms evenly distributed across two buildings on the University of Colorado at Boulder campus were sampled on a single day in November 2010. Surfaces where sampled using sterile, cotton-tipped swabs as described previously?[14],?[15]. As the 12 restrooms were nearly identical in design, we were able to swab the same area at each location between restrooms. In order to characterize tap water communities as a potential source of bacteria, 1 L of faucet water from six of the restrooms (each building having the same water source for each restroom sampled) was collected and filtered through 0.2 ?m bottle top filters (Nalgene, Rochester, NY, USA)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	22132229	NA	NA	NA	Gilberto Flores	CCME	GAZ:United States of America	NA	NA	0	NA	NA	NA	NA	n	NA	Noah Fierer	Flores_restroom_surface_biogeography	pyrosequencing	NA	NA	NA	restroom surface metagenome	3	ENVO:city	Titanium	NA	NA	NA	NA	NA	flores_restroom_surface_biogeography	restroom surface biogeography	256318	CCME	NA	NA	NA	1335	"Genomic DNA was extracted from the swabs and filters using the MO BIO PowerSoil DNA isolation kit following the manufacturer's protocol with the modifications of Fierer?et al.?[14]. A portion of the 16 S rRNA gene spanning the V1_V2 regions was amplified using the primer set (27 F/338R), PCR mixture conditions and thermal cycling conditions described in Fierer?et al.?[15]. PCR amplicons of triplicate reactions for each sample were pooled at approximately equal amounts and pyrosequenced at 454 Life Sciences (Branford, CT, USA) on their GS Junior system. A total of 337,333 high-quality partial 16 S rRNA gene sequences were obtained from 101 of the 120 surface samples collected, averaging approximately 3,340 sequences per sample (ranging from 513_6,771) (Table S1) in 4 GS Junior runs, with the best run containing 116,004 high-quality reads. An additional 16,416 sequences (ranging from 2161_5084 per sample) were generated for five of the six water samples collected for source tracking analysis. Each sample was amplified with a unique barcode to enable multiplexing in the GS Junior runs. The barcoded sequencing reads can be separated by data analysis software providing high confidence in assigning sequencing read to each sample. Sequence data generated as part of this study is available upon request by contacting the corresponding author."	NA	NA	y	NA	NA	ENVO:surface	16S rRNA	Porter	NA	V2	CCME	NA	Soap dispenser	NA	NA	NA	40.01499	NA	NA	Soap dispenser
EKAF6.489460	GACCACTACGAT	YATGCTGCCTCCCGTAGGAGT	jesse.stombaugh@colorado.edu	NA	restroom surface	Toilet_flush_handle	NA	NA	NA	n	12/13/10	restroom surface metagenome	NA	GFL0IUQ	NA	mimarks-survey	NA	NA	NA	NA	0	NA	NA	1624.1	10-Nov	Female	0	NA	NA	NA	NA	NA	NA	NA	NA	NA	NA	"1,swab"	A	NA	Microbial biogeography of public restroom surfaces	NA	NA	-105.27055	y	NA	CCME	NA	NA	NA	"We spend the majority of our lives indoors where we are constantly exposed to bacteria residing on surfaces. However, the diversity of these surface-associated communities is largely unknown. We explored the biogeographical patterns exhibited by bacteria across ten surfaces within each of twelve public restrooms. Using high-throughput barcoded pyrosequencing of the 16S rRNA gene, we identified 19 bacterial phyla across all surfaces. Most sequences belonged to four phyla: Actinobacteria, Bacteriodetes, Firmicutes and Proteobacteria. The communities clustered into three general categories: those found on surfaces associated with toilets, those on the restroom floor, and those found on surfaces routinely touched with hands. On toilet surfaces, gut-associated taxa were more prevalent, suggesting fecal contamination of these surfaces. Floor surfaces were the most diverse of all communities and contained several taxa commonly found in soils. Skin-associated bacteria, especially the Propionibacteriaceae, dominated surfaces routinely touched with our hands. Certain taxa were more common in female than in male restrooms as vagina-associated Lactobacillaceae were widely distributed in female restrooms, likely from urine contamination. Use of the SourceTracker algorithm confirmed many of our taxonomic observations as human skin was the primary source of bacteria on restroom surfaces. Overall, these results demonstrate that restroom surfaces host relatively diverse microbial communities dominated by human-associated bacteria with clear linkages between communities on or in different body sites and those communities found on restroom surfaces. More generally, this work is relevant to the public health field as we show that human associated microbes are commonly found on restroom surfaces suggesting that bacterial pathogens could readily be transmitted between individuals by the touching of surfaces. Furthermore, we demonstrate that we can use high-throughput analyses of bacterial communities to determine sources of bacteria on indoor surfaces, an approach which could be used to track pathogen transmission and test the efficacy of hygiene practices."	ENVO:building	TCAG	Microbial biogeography of public restroom surfaces	NA	0	"Ten surfaces (door handles into and out of the restroom, handles into and out of a restroom stall, faucet handles, soap dispenser, toilet seat, toilet flush handle, floor around the toilet and floor around the sink) in six male and six female restrooms evenly distributed across two buildings on the University of Colorado at Boulder campus were sampled on a single day in November 2010. Surfaces where sampled using sterile, cotton-tipped swabs as described previously?[14],?[15]. As the 12 restrooms were nearly identical in design, we were able to swab the same area at each location between restrooms. In order to characterize tap water communities as a potential source of bacteria, 1 L of faucet water from six of the restrooms (each building having the same water source for each restroom sampled) was collected and filtered through 0.2 ?m bottle top filters (Nalgene, Rochester, NY, USA)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	22132229	NA	NA	NA	Gilberto Flores	CCME	GAZ:United States of America	NA	NA	0	NA	NA	NA	NA	n	NA	Noah Fierer	Flores_restroom_surface_biogeography	pyrosequencing	NA	NA	NA	restroom surface metagenome	1	ENVO:city	Titanium	NA	NA	NA	NA	NA	flores_restroom_surface_biogeography	restroom surface biogeography	256318	CCME	NA	NA	NA	1335	"Genomic DNA was extracted from the swabs and filters using the MO BIO PowerSoil DNA isolation kit following the manufacturer's protocol with the modifications of Fierer?et al.?[14]. A portion of the 16 S rRNA gene spanning the V1_V2 regions was amplified using the primer set (27 F/338R), PCR mixture conditions and thermal cycling conditions described in Fierer?et al.?[15]. PCR amplicons of triplicate reactions for each sample were pooled at approximately equal amounts and pyrosequenced at 454 Life Sciences (Branford, CT, USA) on their GS Junior system. A total of 337,333 high-quality partial 16 S rRNA gene sequences were obtained from 101 of the 120 surface samples collected, averaging approximately 3,340 sequences per sample (ranging from 513_6,771) (Table S1) in 4 GS Junior runs, with the best run containing 116,004 high-quality reads. An additional 16,416 sequences (ranging from 2161_5084 per sample) were generated for five of the six water samples collected for source tracking analysis. Each sample was amplified with a unique barcode to enable multiplexing in the GS Junior runs. The barcoded sequencing reads can be separated by data analysis software providing high confidence in assigning sequencing read to each sample. Sequence data generated as part of this study is available upon request by contacting the corresponding author."	NA	NA	y	NA	NA	ENVO:surface	16S rRNA	Ekeley	NA	V2	CCME	NA	Toilet flush handle	NA	NA	NA	40.01499	NA	NA	Toilet flush handle
EKAM6.489544	GAACTGTATCTC	YATGCTGCCTCCCGTAGGAGT	jesse.stombaugh@colorado.edu	NA	restroom surface	Toilet_flush_handle	NA	NA	NA	n	12/13/10	restroom surface metagenome	NA	GFL0IUQ	NA	mimarks-survey	NA	NA	NA	NA	0	NA	NA	1624.1	10-Nov	Male	0	NA	NA	NA	NA	NA	NA	NA	NA	NA	NA	"1,swab"	A	NA	Microbial biogeography of public restroom surfaces	NA	NA	-105.27055	y	NA	CCME	NA	NA	NA	"We spend the majority of our lives indoors where we are constantly exposed to bacteria residing on surfaces. However, the diversity of these surface-associated communities is largely unknown. We explored the biogeographical patterns exhibited by bacteria across ten surfaces within each of twelve public restrooms. Using high-throughput barcoded pyrosequencing of the 16S rRNA gene, we identified 19 bacterial phyla across all surfaces. Most sequences belonged to four phyla: Actinobacteria, Bacteriodetes, Firmicutes and Proteobacteria. The communities clustered into three general categories: those found on surfaces associated with toilets, those on the restroom floor, and those found on surfaces routinely touched with hands. On toilet surfaces, gut-associated taxa were more prevalent, suggesting fecal contamination of these surfaces. Floor surfaces were the most diverse of all communities and contained several taxa commonly found in soils. Skin-associated bacteria, especially the Propionibacteriaceae, dominated surfaces routinely touched with our hands. Certain taxa were more common in female than in male restrooms as vagina-associated Lactobacillaceae were widely distributed in female restrooms, likely from urine contamination. Use of the SourceTracker algorithm confirmed many of our taxonomic observations as human skin was the primary source of bacteria on restroom surfaces. Overall, these results demonstrate that restroom surfaces host relatively diverse microbial communities dominated by human-associated bacteria with clear linkages between communities on or in different body sites and those communities found on restroom surfaces. More generally, this work is relevant to the public health field as we show that human associated microbes are commonly found on restroom surfaces suggesting that bacterial pathogens could readily be transmitted between individuals by the touching of surfaces. Furthermore, we demonstrate that we can use high-throughput analyses of bacterial communities to determine sources of bacteria on indoor surfaces, an approach which could be used to track pathogen transmission and test the efficacy of hygiene practices."	ENVO:building	TCAG	Microbial biogeography of public restroom surfaces	NA	0	"Ten surfaces (door handles into and out of the restroom, handles into and out of a restroom stall, faucet handles, soap dispenser, toilet seat, toilet flush handle, floor around the toilet and floor around the sink) in six male and six female restrooms evenly distributed across two buildings on the University of Colorado at Boulder campus were sampled on a single day in November 2010. Surfaces where sampled using sterile, cotton-tipped swabs as described previously?[14],?[15]. As the 12 restrooms were nearly identical in design, we were able to swab the same area at each location between restrooms. In order to characterize tap water communities as a potential source of bacteria, 1 L of faucet water from six of the restrooms (each building having the same water source for each restroom sampled) was collected and filtered through 0.2 ?m bottle top filters (Nalgene, Rochester, NY, USA)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	22132229	NA	NA	NA	Gilberto Flores	CCME	GAZ:United States of America	NA	NA	0	NA	NA	NA	NA	n	NA	Noah Fierer	Flores_restroom_surface_biogeography	pyrosequencing	NA	NA	NA	restroom surface metagenome	1	ENVO:city	Titanium	NA	NA	NA	NA	NA	flores_restroom_surface_biogeography	restroom surface biogeography	256318	CCME	NA	NA	NA	1335	"Genomic DNA was extracted from the swabs and filters using the MO BIO PowerSoil DNA isolation kit following the manufacturer's protocol with the modifications of Fierer?et al.?[14]. A portion of the 16 S rRNA gene spanning the V1_V2 regions was amplified using the primer set (27 F/338R), PCR mixture conditions and thermal cycling conditions described in Fierer?et al.?[15]. PCR amplicons of triplicate reactions for each sample were pooled at approximately equal amounts and pyrosequenced at 454 Life Sciences (Branford, CT, USA) on their GS Junior system. A total of 337,333 high-quality partial 16 S rRNA gene sequences were obtained from 101 of the 120 surface samples collected, averaging approximately 3,340 sequences per sample (ranging from 513_6,771) (Table S1) in 4 GS Junior runs, with the best run containing 116,004 high-quality reads. An additional 16,416 sequences (ranging from 2161_5084 per sample) were generated for five of the six water samples collected for source tracking analysis. Each sample was amplified with a unique barcode to enable multiplexing in the GS Junior runs. The barcoded sequencing reads can be separated by data analysis software providing high confidence in assigning sequencing read to each sample. Sequence data generated as part of this study is available upon request by contacting the corresponding author."	NA	NA	y	NA	NA	ENVO:surface	16S rRNA	Ekeley	NA	V2	CCME	NA	Toilet flush handle	NA	NA	NA	40.01499	NA	NA	Toilet flush handle
EKBM6.489504	GACTCGAATCGT	YATGCTGCCTCCCGTAGGAGT	jesse.stombaugh@colorado.edu	NA	restroom surface	Toilet_flush_handle	NA	NA	NA	n	12/13/10	restroom surface metagenome	NA	GFL0IUQ	NA	mimarks-survey	NA	NA	NA	NA	0	NA	NA	1624.1	10-Nov	Male	0	NA	NA	NA	NA	NA	NA	NA	NA	NA	NA	"1,swab"	B	NA	Microbial biogeography of public restroom surfaces	NA	NA	-105.27055	y	NA	CCME	NA	NA	NA	"We spend the majority of our lives indoors where we are constantly exposed to bacteria residing on surfaces. However, the diversity of these surface-associated communities is largely unknown. We explored the biogeographical patterns exhibited by bacteria across ten surfaces within each of twelve public restrooms. Using high-throughput barcoded pyrosequencing of the 16S rRNA gene, we identified 19 bacterial phyla across all surfaces. Most sequences belonged to four phyla: Actinobacteria, Bacteriodetes, Firmicutes and Proteobacteria. The communities clustered into three general categories: those found on surfaces associated with toilets, those on the restroom floor, and those found on surfaces routinely touched with hands. On toilet surfaces, gut-associated taxa were more prevalent, suggesting fecal contamination of these surfaces. Floor surfaces were the most diverse of all communities and contained several taxa commonly found in soils. Skin-associated bacteria, especially the Propionibacteriaceae, dominated surfaces routinely touched with our hands. Certain taxa were more common in female than in male restrooms as vagina-associated Lactobacillaceae were widely distributed in female restrooms, likely from urine contamination. Use of the SourceTracker algorithm confirmed many of our taxonomic observations as human skin was the primary source of bacteria on restroom surfaces. Overall, these results demonstrate that restroom surfaces host relatively diverse microbial communities dominated by human-associated bacteria with clear linkages between communities on or in different body sites and those communities found on restroom surfaces. More generally, this work is relevant to the public health field as we show that human associated microbes are commonly found on restroom surfaces suggesting that bacterial pathogens could readily be transmitted between individuals by the touching of surfaces. Furthermore, we demonstrate that we can use high-throughput analyses of bacterial communities to determine sources of bacteria on indoor surfaces, an approach which could be used to track pathogen transmission and test the efficacy of hygiene practices."	ENVO:building	TCAG	Microbial biogeography of public restroom surfaces	NA	0	"Ten surfaces (door handles into and out of the restroom, handles into and out of a restroom stall, faucet handles, soap dispenser, toilet seat, toilet flush handle, floor around the toilet and floor around the sink) in six male and six female restrooms evenly distributed across two buildings on the University of Colorado at Boulder campus were sampled on a single day in November 2010. Surfaces where sampled using sterile, cotton-tipped swabs as described previously?[14],?[15]. As the 12 restrooms were nearly identical in design, we were able to swab the same area at each location between restrooms. In order to characterize tap water communities as a potential source of bacteria, 1 L of faucet water from six of the restrooms (each building having the same water source for each restroom sampled) was collected and filtered through 0.2 ?m bottle top filters (Nalgene, Rochester, NY, USA)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	22132229	NA	NA	NA	Gilberto Flores	CCME	GAZ:United States of America	NA	NA	0	NA	NA	NA	NA	n	NA	Noah Fierer	Flores_restroom_surface_biogeography	pyrosequencing	NA	NA	NA	restroom surface metagenome	2	ENVO:city	Titanium	NA	NA	NA	NA	NA	flores_restroom_surface_biogeography	restroom surface biogeography	256318	CCME	NA	NA	NA	1335	"Genomic DNA was extracted from the swabs and filters using the MO BIO PowerSoil DNA isolation kit following the manufacturer's protocol with the modifications of Fierer?et al.?[14]. A portion of the 16 S rRNA gene spanning the V1_V2 regions was amplified using the primer set (27 F/338R), PCR mixture conditions and thermal cycling conditions described in Fierer?et al.?[15]. PCR amplicons of triplicate reactions for each sample were pooled at approximately equal amounts and pyrosequenced at 454 Life Sciences (Branford, CT, USA) on their GS Junior system. A total of 337,333 high-quality partial 16 S rRNA gene sequences were obtained from 101 of the 120 surface samples collected, averaging approximately 3,340 sequences per sample (ranging from 513_6,771) (Table S1) in 4 GS Junior runs, with the best run containing 116,004 high-quality reads. An additional 16,416 sequences (ranging from 2161_5084 per sample) were generated for five of the six water samples collected for source tracking analysis. Each sample was amplified with a unique barcode to enable multiplexing in the GS Junior runs. The barcoded sequencing reads can be separated by data analysis software providing high confidence in assigning sequencing read to each sample. Sequence data generated as part of this study is available upon request by contacting the corresponding author."	NA	NA	y	NA	NA	ENVO:surface	16S rRNA	Ekeley	NA	V2	CCME	NA	Toilet flush handle	NA	NA	NA	40.01499	NA	NA	Toilet flush handle
EKCF6.489456	GATGATCGCCGA	YATGCTGCCTCCCGTAGGAGT	jesse.stombaugh@colorado.edu	NA	restroom surface	Toilet_flush_handle	NA	NA	NA	n	12/13/10	restroom surface metagenome	NA	GFL0IUQ	NA	mimarks-survey	NA	NA	NA	NA	0	NA	NA	1624.1	10-Nov	Female	0	NA	NA	NA	NA	NA	NA	NA	NA	NA	NA	"1,swab"	C	NA	Microbial biogeography of public restroom surfaces	NA	NA	-105.27055	y	NA	CCME	NA	NA	NA	"We spend the majority of our lives indoors where we are constantly exposed to bacteria residing on surfaces. However, the diversity of these surface-associated communities is largely unknown. We explored the biogeographical patterns exhibited by bacteria across ten surfaces within each of twelve public restrooms. Using high-throughput barcoded pyrosequencing of the 16S rRNA gene, we identified 19 bacterial phyla across all surfaces. Most sequences belonged to four phyla: Actinobacteria, Bacteriodetes, Firmicutes and Proteobacteria. The communities clustered into three general categories: those found on surfaces associated with toilets, those on the restroom floor, and those found on surfaces routinely touched with hands. On toilet surfaces, gut-associated taxa were more prevalent, suggesting fecal contamination of these surfaces. Floor surfaces were the most diverse of all communities and contained several taxa commonly found in soils. Skin-associated bacteria, especially the Propionibacteriaceae, dominated surfaces routinely touched with our hands. Certain taxa were more common in female than in male restrooms as vagina-associated Lactobacillaceae were widely distributed in female restrooms, likely from urine contamination. Use of the SourceTracker algorithm confirmed many of our taxonomic observations as human skin was the primary source of bacteria on restroom surfaces. Overall, these results demonstrate that restroom surfaces host relatively diverse microbial communities dominated by human-associated bacteria with clear linkages between communities on or in different body sites and those communities found on restroom surfaces. More generally, this work is relevant to the public health field as we show that human associated microbes are commonly found on restroom surfaces suggesting that bacterial pathogens could readily be transmitted between individuals by the touching of surfaces. Furthermore, we demonstrate that we can use high-throughput analyses of bacterial communities to determine sources of bacteria on indoor surfaces, an approach which could be used to track pathogen transmission and test the efficacy of hygiene practices."	ENVO:building	TCAG	Microbial biogeography of public restroom surfaces	NA	0	"Ten surfaces (door handles into and out of the restroom, handles into and out of a restroom stall, faucet handles, soap dispenser, toilet seat, toilet flush handle, floor around the toilet and floor around the sink) in six male and six female restrooms evenly distributed across two buildings on the University of Colorado at Boulder campus were sampled on a single day in November 2010. Surfaces where sampled using sterile, cotton-tipped swabs as described previously?[14],?[15]. As the 12 restrooms were nearly identical in design, we were able to swab the same area at each location between restrooms. In order to characterize tap water communities as a potential source of bacteria, 1 L of faucet water from six of the restrooms (each building having the same water source for each restroom sampled) was collected and filtered through 0.2 ?m bottle top filters (Nalgene, Rochester, NY, USA)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	22132229	NA	NA	NA	Gilberto Flores	CCME	GAZ:United States of America	NA	NA	0	NA	NA	NA	NA	n	NA	Noah Fierer	Flores_restroom_surface_biogeography	pyrosequencing	NA	NA	NA	restroom surface metagenome	3	ENVO:city	Titanium	NA	NA	NA	NA	NA	flores_restroom_surface_biogeography	restroom surface biogeography	256318	CCME	NA	NA	NA	1335	"Genomic DNA was extracted from the swabs and filters using the MO BIO PowerSoil DNA isolation kit following the manufacturer's protocol with the modifications of Fierer?et al.?[14]. A portion of the 16 S rRNA gene spanning the V1_V2 regions was amplified using the primer set (27 F/338R), PCR mixture conditions and thermal cycling conditions described in Fierer?et al.?[15]. PCR amplicons of triplicate reactions for each sample were pooled at approximately equal amounts and pyrosequenced at 454 Life Sciences (Branford, CT, USA) on their GS Junior system. A total of 337,333 high-quality partial 16 S rRNA gene sequences were obtained from 101 of the 120 surface samples collected, averaging approximately 3,340 sequences per sample (ranging from 513_6,771) (Table S1) in 4 GS Junior runs, with the best run containing 116,004 high-quality reads. An additional 16,416 sequences (ranging from 2161_5084 per sample) were generated for five of the six water samples collected for source tracking analysis. Each sample was amplified with a unique barcode to enable multiplexing in the GS Junior runs. The barcoded sequencing reads can be separated by data analysis software providing high confidence in assigning sequencing read to each sample. Sequence data generated as part of this study is available upon request by contacting the corresponding author."	NA	NA	y	NA	NA	ENVO:surface	16S rRNA	Ekeley	NA	V2	CCME	NA	Toilet flush handle	NA	NA	NA	40.01499	NA	NA	Toilet flush handle
EKCM6.489475	GATAGCTGTCTT	YATGCTGCCTCCCGTAGGAGT	jesse.stombaugh@colorado.edu	NA	restroom surface	Toilet_flush_handle	NA	NA	NA	n	12/13/10	restroom surface metagenome	NA	GFL0IUQ	NA	mimarks-survey	NA	NA	NA	NA	0	NA	NA	1624.1	10-Nov	Male	0	NA	NA	NA	NA	NA	NA	NA	NA	NA	NA	"1,swab"	C	NA	Microbial biogeography of public restroom surfaces	NA	NA	-105.27055	y	NA	CCME	NA	NA	NA	"We spend the majority of our lives indoors where we are constantly exposed to bacteria residing on surfaces. However, the diversity of these surface-associated communities is largely unknown. We explored the biogeographical patterns exhibited by bacteria across ten surfaces within each of twelve public restrooms. Using high-throughput barcoded pyrosequencing of the 16S rRNA gene, we identified 19 bacterial phyla across all surfaces. Most sequences belonged to four phyla: Actinobacteria, Bacteriodetes, Firmicutes and Proteobacteria. The communities clustered into three general categories: those found on surfaces associated with toilets, those on the restroom floor, and those found on surfaces routinely touched with hands. On toilet surfaces, gut-associated taxa were more prevalent, suggesting fecal contamination of these surfaces. Floor surfaces were the most diverse of all communities and contained several taxa commonly found in soils. Skin-associated bacteria, especially the Propionibacteriaceae, dominated surfaces routinely touched with our hands. Certain taxa were more common in female than in male restrooms as vagina-associated Lactobacillaceae were widely distributed in female restrooms, likely from urine contamination. Use of the SourceTracker algorithm confirmed many of our taxonomic observations as human skin was the primary source of bacteria on restroom surfaces. Overall, these results demonstrate that restroom surfaces host relatively diverse microbial communities dominated by human-associated bacteria with clear linkages between communities on or in different body sites and those communities found on restroom surfaces. More generally, this work is relevant to the public health field as we show that human associated microbes are commonly found on restroom surfaces suggesting that bacterial pathogens could readily be transmitted between individuals by the touching of surfaces. Furthermore, we demonstrate that we can use high-throughput analyses of bacterial communities to determine sources of bacteria on indoor surfaces, an approach which could be used to track pathogen transmission and test the efficacy of hygiene practices."	ENVO:building	TCAG	Microbial biogeography of public restroom surfaces	NA	0	"Ten surfaces (door handles into and out of the restroom, handles into and out of a restroom stall, faucet handles, soap dispenser, toilet seat, toilet flush handle, floor around the toilet and floor around the sink) in six male and six female restrooms evenly distributed across two buildings on the University of Colorado at Boulder campus were sampled on a single day in November 2010. Surfaces where sampled using sterile, cotton-tipped swabs as described previously?[14],?[15]. As the 12 restrooms were nearly identical in design, we were able to swab the same area at each location between restrooms. In order to characterize tap water communities as a potential source of bacteria, 1 L of faucet water from six of the restrooms (each building having the same water source for each restroom sampled) was collected and filtered through 0.2 ?m bottle top filters (Nalgene, Rochester, NY, USA)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	22132229	NA	NA	NA	Gilberto Flores	CCME	GAZ:United States of America	NA	NA	0	NA	NA	NA	NA	n	NA	Noah Fierer	Flores_restroom_surface_biogeography	pyrosequencing	NA	NA	NA	restroom surface metagenome	3	ENVO:city	Titanium	NA	NA	NA	NA	NA	flores_restroom_surface_biogeography	restroom surface biogeography	256318	CCME	NA	NA	NA	1335	"Genomic DNA was extracted from the swabs and filters using the MO BIO PowerSoil DNA isolation kit following the manufacturer's protocol with the modifications of Fierer?et al.?[14]. A portion of the 16 S rRNA gene spanning the V1_V2 regions was amplified using the primer set (27 F/338R), PCR mixture conditions and thermal cycling conditions described in Fierer?et al.?[15]. PCR amplicons of triplicate reactions for each sample were pooled at approximately equal amounts and pyrosequenced at 454 Life Sciences (Branford, CT, USA) on their GS Junior system. A total of 337,333 high-quality partial 16 S rRNA gene sequences were obtained from 101 of the 120 surface samples collected, averaging approximately 3,340 sequences per sample (ranging from 513_6,771) (Table S1) in 4 GS Junior runs, with the best run containing 116,004 high-quality reads. An additional 16,416 sequences (ranging from 2161_5084 per sample) were generated for five of the six water samples collected for source tracking analysis. Each sample was amplified with a unique barcode to enable multiplexing in the GS Junior runs. The barcoded sequencing reads can be separated by data analysis software providing high confidence in assigning sequencing read to each sample. Sequence data generated as part of this study is available upon request by contacting the corresponding author."	NA	NA	y	NA	NA	ENVO:surface	16S rRNA	Ekeley	NA	V2	CCME	NA	Toilet flush handle	NA	NA	NA	40.01499	NA	NA	Toilet flush handle
PTAF6.489492	GCATTGCGTGAG	YATGCTGCCTCCCGTAGGAGT	jesse.stombaugh@colorado.edu	NA	restroom surface	Toilet_flush_handle	NA	NA	NA	n	12/13/10	restroom surface metagenome	NA	GFL0IUQ	NA	mimarks-survey	NA	NA	NA	NA	0	NA	NA	1624.1	10-Nov	Female	0	NA	NA	NA	NA	NA	NA	NA	NA	NA	NA	"1,swab"	A	NA	Microbial biogeography of public restroom surfaces	NA	NA	-105.27055	y	NA	CCME	NA	NA	NA	"We spend the majority of our lives indoors where we are constantly exposed to bacteria residing on surfaces. However, the diversity of these surface-associated communities is largely unknown. We explored the biogeographical patterns exhibited by bacteria across ten surfaces within each of twelve public restrooms. Using high-throughput barcoded pyrosequencing of the 16S rRNA gene, we identified 19 bacterial phyla across all surfaces. Most sequences belonged to four phyla: Actinobacteria, Bacteriodetes, Firmicutes and Proteobacteria. The communities clustered into three general categories: those found on surfaces associated with toilets, those on the restroom floor, and those found on surfaces routinely touched with hands. On toilet surfaces, gut-associated taxa were more prevalent, suggesting fecal contamination of these surfaces. Floor surfaces were the most diverse of all communities and contained several taxa commonly found in soils. Skin-associated bacteria, especially the Propionibacteriaceae, dominated surfaces routinely touched with our hands. Certain taxa were more common in female than in male restrooms as vagina-associated Lactobacillaceae were widely distributed in female restrooms, likely from urine contamination. Use of the SourceTracker algorithm confirmed many of our taxonomic observations as human skin was the primary source of bacteria on restroom surfaces. Overall, these results demonstrate that restroom surfaces host relatively diverse microbial communities dominated by human-associated bacteria with clear linkages between communities on or in different body sites and those communities found on restroom surfaces. More generally, this work is relevant to the public health field as we show that human associated microbes are commonly found on restroom surfaces suggesting that bacterial pathogens could readily be transmitted between individuals by the touching of surfaces. Furthermore, we demonstrate that we can use high-throughput analyses of bacterial communities to determine sources of bacteria on indoor surfaces, an approach which could be used to track pathogen transmission and test the efficacy of hygiene practices."	ENVO:building	TCAG	Microbial biogeography of public restroom surfaces	NA	0	"Ten surfaces (door handles into and out of the restroom, handles into and out of a restroom stall, faucet handles, soap dispenser, toilet seat, toilet flush handle, floor around the toilet and floor around the sink) in six male and six female restrooms evenly distributed across two buildings on the University of Colorado at Boulder campus were sampled on a single day in November 2010. Surfaces where sampled using sterile, cotton-tipped swabs as described previously?[14],?[15]. As the 12 restrooms were nearly identical in design, we were able to swab the same area at each location between restrooms. In order to characterize tap water communities as a potential source of bacteria, 1 L of faucet water from six of the restrooms (each building having the same water source for each restroom sampled) was collected and filtered through 0.2 ?m bottle top filters (Nalgene, Rochester, NY, USA)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	22132229	NA	NA	NA	Gilberto Flores	CCME	GAZ:United States of America	NA	NA	0	NA	NA	NA	NA	n	NA	Noah Fierer	Flores_restroom_surface_biogeography	pyrosequencing	NA	NA	NA	restroom surface metagenome	1	ENVO:city	Titanium	NA	NA	NA	NA	NA	flores_restroom_surface_biogeography	restroom surface biogeography	256318	CCME	NA	NA	NA	1335	"Genomic DNA was extracted from the swabs and filters using the MO BIO PowerSoil DNA isolation kit following the manufacturer's protocol with the modifications of Fierer?et al.?[14]. A portion of the 16 S rRNA gene spanning the V1_V2 regions was amplified using the primer set (27 F/338R), PCR mixture conditions and thermal cycling conditions described in Fierer?et al.?[15]. PCR amplicons of triplicate reactions for each sample were pooled at approximately equal amounts and pyrosequenced at 454 Life Sciences (Branford, CT, USA) on their GS Junior system. A total of 337,333 high-quality partial 16 S rRNA gene sequences were obtained from 101 of the 120 surface samples collected, averaging approximately 3,340 sequences per sample (ranging from 513_6,771) (Table S1) in 4 GS Junior runs, with the best run containing 116,004 high-quality reads. An additional 16,416 sequences (ranging from 2161_5084 per sample) were generated for five of the six water samples collected for source tracking analysis. Each sample was amplified with a unique barcode to enable multiplexing in the GS Junior runs. The barcoded sequencing reads can be separated by data analysis software providing high confidence in assigning sequencing read to each sample. Sequence data generated as part of this study is available upon request by contacting the corresponding author."	NA	NA	y	NA	NA	ENVO:surface	16S rRNA	Porter	NA	V2	CCME	NA	Toilet flush handle	NA	NA	NA	40.01499	NA	NA	Toilet flush handle
PTAM6.489451	GCAGCACGTTGA	YATGCTGCCTCCCGTAGGAGT	jesse.stombaugh@colorado.edu	NA	restroom surface	Toilet_flush_handle	NA	NA	NA	n	12/13/10	restroom surface metagenome	NA	GFL0IUQ	NA	mimarks-survey	NA	NA	NA	NA	0	NA	NA	1624.1	10-Nov	Male	0	NA	NA	NA	NA	NA	NA	NA	NA	NA	NA	"1,swab"	A	NA	Microbial biogeography of public restroom surfaces	NA	NA	-105.27055	y	NA	CCME	NA	NA	NA	"We spend the majority of our lives indoors where we are constantly exposed to bacteria residing on surfaces. However, the diversity of these surface-associated communities is largely unknown. We explored the biogeographical patterns exhibited by bacteria across ten surfaces within each of twelve public restrooms. Using high-throughput barcoded pyrosequencing of the 16S rRNA gene, we identified 19 bacterial phyla across all surfaces. Most sequences belonged to four phyla: Actinobacteria, Bacteriodetes, Firmicutes and Proteobacteria. The communities clustered into three general categories: those found on surfaces associated with toilets, those on the restroom floor, and those found on surfaces routinely touched with hands. On toilet surfaces, gut-associated taxa were more prevalent, suggesting fecal contamination of these surfaces. Floor surfaces were the most diverse of all communities and contained several taxa commonly found in soils. Skin-associated bacteria, especially the Propionibacteriaceae, dominated surfaces routinely touched with our hands. Certain taxa were more common in female than in male restrooms as vagina-associated Lactobacillaceae were widely distributed in female restrooms, likely from urine contamination. Use of the SourceTracker algorithm confirmed many of our taxonomic observations as human skin was the primary source of bacteria on restroom surfaces. Overall, these results demonstrate that restroom surfaces host relatively diverse microbial communities dominated by human-associated bacteria with clear linkages between communities on or in different body sites and those communities found on restroom surfaces. More generally, this work is relevant to the public health field as we show that human associated microbes are commonly found on restroom surfaces suggesting that bacterial pathogens could readily be transmitted between individuals by the touching of surfaces. Furthermore, we demonstrate that we can use high-throughput analyses of bacterial communities to determine sources of bacteria on indoor surfaces, an approach which could be used to track pathogen transmission and test the efficacy of hygiene practices."	ENVO:building	TCAG	Microbial biogeography of public restroom surfaces	NA	0	"Ten surfaces (door handles into and out of the restroom, handles into and out of a restroom stall, faucet handles, soap dispenser, toilet seat, toilet flush handle, floor around the toilet and floor around the sink) in six male and six female restrooms evenly distributed across two buildings on the University of Colorado at Boulder campus were sampled on a single day in November 2010. Surfaces where sampled using sterile, cotton-tipped swabs as described previously?[14],?[15]. As the 12 restrooms were nearly identical in design, we were able to swab the same area at each location between restrooms. In order to characterize tap water communities as a potential source of bacteria, 1 L of faucet water from six of the restrooms (each building having the same water source for each restroom sampled) was collected and filtered through 0.2 ?m bottle top filters (Nalgene, Rochester, NY, USA)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	22132229	NA	NA	NA	Gilberto Flores	CCME	GAZ:United States of America	NA	NA	0	NA	NA	NA	NA	n	NA	Noah Fierer	Flores_restroom_surface_biogeography	pyrosequencing	NA	NA	NA	restroom surface metagenome	1	ENVO:city	Titanium	NA	NA	NA	NA	NA	flores_restroom_surface_biogeography	restroom surface biogeography	256318	CCME	NA	NA	NA	1335	"Genomic DNA was extracted from the swabs and filters using the MO BIO PowerSoil DNA isolation kit following the manufacturer's protocol with the modifications of Fierer?et al.?[14]. A portion of the 16 S rRNA gene spanning the V1_V2 regions was amplified using the primer set (27 F/338R), PCR mixture conditions and thermal cycling conditions described in Fierer?et al.?[15]. PCR amplicons of triplicate reactions for each sample were pooled at approximately equal amounts and pyrosequenced at 454 Life Sciences (Branford, CT, USA) on their GS Junior system. A total of 337,333 high-quality partial 16 S rRNA gene sequences were obtained from 101 of the 120 surface samples collected, averaging approximately 3,340 sequences per sample (ranging from 513_6,771) (Table S1) in 4 GS Junior runs, with the best run containing 116,004 high-quality reads. An additional 16,416 sequences (ranging from 2161_5084 per sample) were generated for five of the six water samples collected for source tracking analysis. Each sample was amplified with a unique barcode to enable multiplexing in the GS Junior runs. The barcoded sequencing reads can be separated by data analysis software providing high confidence in assigning sequencing read to each sample. Sequence data generated as part of this study is available upon request by contacting the corresponding author."	NA	NA	y	NA	NA	ENVO:surface	16S rRNA	Porter	NA	V2	CCME	NA	Toilet flush handle	NA	NA	NA	40.01499	NA	NA	Toilet flush handle
PTBM6.489476	GCGTTACACACA	YATGCTGCCTCCCGTAGGAGT	jesse.stombaugh@colorado.edu	NA	restroom surface	Toilet_flush_handle	NA	NA	NA	n	12/13/10	restroom surface metagenome	NA	GFL0IUQ	NA	mimarks-survey	NA	NA	NA	NA	0	NA	NA	1624.1	10-Nov	Male	0	NA	NA	NA	NA	NA	NA	NA	NA	NA	NA	"1,swab"	B	NA	Microbial biogeography of public restroom surfaces	NA	NA	-105.27055	y	NA	CCME	NA	NA	NA	"We spend the majority of our lives indoors where we are constantly exposed to bacteria residing on surfaces. However, the diversity of these surface-associated communities is largely unknown. We explored the biogeographical patterns exhibited by bacteria across ten surfaces within each of twelve public restrooms. Using high-throughput barcoded pyrosequencing of the 16S rRNA gene, we identified 19 bacterial phyla across all surfaces. Most sequences belonged to four phyla: Actinobacteria, Bacteriodetes, Firmicutes and Proteobacteria. The communities clustered into three general categories: those found on surfaces associated with toilets, those on the restroom floor, and those found on surfaces routinely touched with hands. On toilet surfaces, gut-associated taxa were more prevalent, suggesting fecal contamination of these surfaces. Floor surfaces were the most diverse of all communities and contained several taxa commonly found in soils. Skin-associated bacteria, especially the Propionibacteriaceae, dominated surfaces routinely touched with our hands. Certain taxa were more common in female than in male restrooms as vagina-associated Lactobacillaceae were widely distributed in female restrooms, likely from urine contamination. Use of the SourceTracker algorithm confirmed many of our taxonomic observations as human skin was the primary source of bacteria on restroom surfaces. Overall, these results demonstrate that restroom surfaces host relatively diverse microbial communities dominated by human-associated bacteria with clear linkages between communities on or in different body sites and those communities found on restroom surfaces. More generally, this work is relevant to the public health field as we show that human associated microbes are commonly found on restroom surfaces suggesting that bacterial pathogens could readily be transmitted between individuals by the touching of surfaces. Furthermore, we demonstrate that we can use high-throughput analyses of bacterial communities to determine sources of bacteria on indoor surfaces, an approach which could be used to track pathogen transmission and test the efficacy of hygiene practices."	ENVO:building	TCAG	Microbial biogeography of public restroom surfaces	NA	0	"Ten surfaces (door handles into and out of the restroom, handles into and out of a restroom stall, faucet handles, soap dispenser, toilet seat, toilet flush handle, floor around the toilet and floor around the sink) in six male and six female restrooms evenly distributed across two buildings on the University of Colorado at Boulder campus were sampled on a single day in November 2010. Surfaces where sampled using sterile, cotton-tipped swabs as described previously?[14],?[15]. As the 12 restrooms were nearly identical in design, we were able to swab the same area at each location between restrooms. In order to characterize tap water communities as a potential source of bacteria, 1 L of faucet water from six of the restrooms (each building having the same water source for each restroom sampled) was collected and filtered through 0.2 ?m bottle top filters (Nalgene, Rochester, NY, USA)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	22132229	NA	NA	NA	Gilberto Flores	CCME	GAZ:United States of America	NA	NA	0	NA	NA	NA	NA	n	NA	Noah Fierer	Flores_restroom_surface_biogeography	pyrosequencing	NA	NA	NA	restroom surface metagenome	2	ENVO:city	Titanium	NA	NA	NA	NA	NA	flores_restroom_surface_biogeography	restroom surface biogeography	256318	CCME	NA	NA	NA	1335	"Genomic DNA was extracted from the swabs and filters using the MO BIO PowerSoil DNA isolation kit following the manufacturer's protocol with the modifications of Fierer?et al.?[14]. A portion of the 16 S rRNA gene spanning the V1_V2 regions was amplified using the primer set (27 F/338R), PCR mixture conditions and thermal cycling conditions described in Fierer?et al.?[15]. PCR amplicons of triplicate reactions for each sample were pooled at approximately equal amounts and pyrosequenced at 454 Life Sciences (Branford, CT, USA) on their GS Junior system. A total of 337,333 high-quality partial 16 S rRNA gene sequences were obtained from 101 of the 120 surface samples collected, averaging approximately 3,340 sequences per sample (ranging from 513_6,771) (Table S1) in 4 GS Junior runs, with the best run containing 116,004 high-quality reads. An additional 16,416 sequences (ranging from 2161_5084 per sample) were generated for five of the six water samples collected for source tracking analysis. Each sample was amplified with a unique barcode to enable multiplexing in the GS Junior runs. The barcoded sequencing reads can be separated by data analysis software providing high confidence in assigning sequencing read to each sample. Sequence data generated as part of this study is available upon request by contacting the corresponding author."	NA	NA	y	NA	NA	ENVO:surface	16S rRNA	Porter	NA	V2	CCME	NA	Toilet flush handle	NA	NA	NA	40.01499	NA	NA	Toilet flush handle
PTCF6.489533	ATCAGGCGTGTG	YATGCTGCCTCCCGTAGGAGT	jesse.stombaugh@colorado.edu	NA	restroom surface	Toilet_flush_handle	NA	NA	NA	n	12/13/10	restroom surface metagenome	NA	GFL0IUQ	NA	mimarks-survey	NA	NA	NA	NA	0	NA	NA	1624.1	10-Nov	Female	0	NA	NA	NA	NA	NA	NA	NA	NA	NA	NA	"1,swab"	C	NA	Microbial biogeography of public restroom surfaces	NA	NA	-105.27055	y	NA	CCME	NA	NA	NA	"We spend the majority of our lives indoors where we are constantly exposed to bacteria residing on surfaces. However, the diversity of these surface-associated communities is largely unknown. We explored the biogeographical patterns exhibited by bacteria across ten surfaces within each of twelve public restrooms. Using high-throughput barcoded pyrosequencing of the 16S rRNA gene, we identified 19 bacterial phyla across all surfaces. Most sequences belonged to four phyla: Actinobacteria, Bacteriodetes, Firmicutes and Proteobacteria. The communities clustered into three general categories: those found on surfaces associated with toilets, those on the restroom floor, and those found on surfaces routinely touched with hands. On toilet surfaces, gut-associated taxa were more prevalent, suggesting fecal contamination of these surfaces. Floor surfaces were the most diverse of all communities and contained several taxa commonly found in soils. Skin-associated bacteria, especially the Propionibacteriaceae, dominated surfaces routinely touched with our hands. Certain taxa were more common in female than in male restrooms as vagina-associated Lactobacillaceae were widely distributed in female restrooms, likely from urine contamination. Use of the SourceTracker algorithm confirmed many of our taxonomic observations as human skin was the primary source of bacteria on restroom surfaces. Overall, these results demonstrate that restroom surfaces host relatively diverse microbial communities dominated by human-associated bacteria with clear linkages between communities on or in different body sites and those communities found on restroom surfaces. More generally, this work is relevant to the public health field as we show that human associated microbes are commonly found on restroom surfaces suggesting that bacterial pathogens could readily be transmitted between individuals by the touching of surfaces. Furthermore, we demonstrate that we can use high-throughput analyses of bacterial communities to determine sources of bacteria on indoor surfaces, an approach which could be used to track pathogen transmission and test the efficacy of hygiene practices."	ENVO:building	TCAG	Microbial biogeography of public restroom surfaces	NA	0	"Ten surfaces (door handles into and out of the restroom, handles into and out of a restroom stall, faucet handles, soap dispenser, toilet seat, toilet flush handle, floor around the toilet and floor around the sink) in six male and six female restrooms evenly distributed across two buildings on the University of Colorado at Boulder campus were sampled on a single day in November 2010. Surfaces where sampled using sterile, cotton-tipped swabs as described previously?[14],?[15]. As the 12 restrooms were nearly identical in design, we were able to swab the same area at each location between restrooms. In order to characterize tap water communities as a potential source of bacteria, 1 L of faucet water from six of the restrooms (each building having the same water source for each restroom sampled) was collected and filtered through 0.2 ?m bottle top filters (Nalgene, Rochester, NY, USA)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	22132229	NA	NA	NA	Gilberto Flores	CCME	GAZ:United States of America	NA	NA	0	NA	NA	NA	NA	n	NA	Noah Fierer	Flores_restroom_surface_biogeography	pyrosequencing	NA	NA	NA	restroom surface metagenome	3	ENVO:city	Titanium	NA	NA	NA	NA	NA	flores_restroom_surface_biogeography	restroom surface biogeography	256318	CCME	NA	NA	NA	1335	"Genomic DNA was extracted from the swabs and filters using the MO BIO PowerSoil DNA isolation kit following the manufacturer's protocol with the modifications of Fierer?et al.?[14]. A portion of the 16 S rRNA gene spanning the V1_V2 regions was amplified using the primer set (27 F/338R), PCR mixture conditions and thermal cycling conditions described in Fierer?et al.?[15]. PCR amplicons of triplicate reactions for each sample were pooled at approximately equal amounts and pyrosequenced at 454 Life Sciences (Branford, CT, USA) on their GS Junior system. A total of 337,333 high-quality partial 16 S rRNA gene sequences were obtained from 101 of the 120 surface samples collected, averaging approximately 3,340 sequences per sample (ranging from 513_6,771) (Table S1) in 4 GS Junior runs, with the best run containing 116,004 high-quality reads. An additional 16,416 sequences (ranging from 2161_5084 per sample) were generated for five of the six water samples collected for source tracking analysis. Each sample was amplified with a unique barcode to enable multiplexing in the GS Junior runs. The barcoded sequencing reads can be separated by data analysis software providing high confidence in assigning sequencing read to each sample. Sequence data generated as part of this study is available upon request by contacting the corresponding author."	NA	NA	y	NA	NA	ENVO:surface	16S rRNA	Porter	NA	V2	CCME	NA	Toilet flush handle	NA	NA	NA	40.01499	NA	NA	Toilet flush handle
PTCM6.489574	ATACAGAGCTCC	YATGCTGCCTCCCGTAGGAGT	jesse.stombaugh@colorado.edu	NA	restroom surface	Toilet_flush_handle	NA	NA	NA	n	12/13/10	restroom surface metagenome	NA	GFL0IUQ	NA	mimarks-survey	NA	NA	NA	NA	0	NA	NA	1624.1	10-Nov	Male	0	NA	NA	NA	NA	NA	NA	NA	NA	NA	NA	"1,swab"	C	NA	Microbial biogeography of public restroom surfaces	NA	NA	-105.27055	y	NA	CCME	NA	NA	NA	"We spend the majority of our lives indoors where we are constantly exposed to bacteria residing on surfaces. However, the diversity of these surface-associated communities is largely unknown. We explored the biogeographical patterns exhibited by bacteria across ten surfaces within each of twelve public restrooms. Using high-throughput barcoded pyrosequencing of the 16S rRNA gene, we identified 19 bacterial phyla across all surfaces. Most sequences belonged to four phyla: Actinobacteria, Bacteriodetes, Firmicutes and Proteobacteria. The communities clustered into three general categories: those found on surfaces associated with toilets, those on the restroom floor, and those found on surfaces routinely touched with hands. On toilet surfaces, gut-associated taxa were more prevalent, suggesting fecal contamination of these surfaces. Floor surfaces were the most diverse of all communities and contained several taxa commonly found in soils. Skin-associated bacteria, especially the Propionibacteriaceae, dominated surfaces routinely touched with our hands. Certain taxa were more common in female than in male restrooms as vagina-associated Lactobacillaceae were widely distributed in female restrooms, likely from urine contamination. Use of the SourceTracker algorithm confirmed many of our taxonomic observations as human skin was the primary source of bacteria on restroom surfaces. Overall, these results demonstrate that restroom surfaces host relatively diverse microbial communities dominated by human-associated bacteria with clear linkages between communities on or in different body sites and those communities found on restroom surfaces. More generally, this work is relevant to the public health field as we show that human associated microbes are commonly found on restroom surfaces suggesting that bacterial pathogens could readily be transmitted between individuals by the touching of surfaces. Furthermore, we demonstrate that we can use high-throughput analyses of bacterial communities to determine sources of bacteria on indoor surfaces, an approach which could be used to track pathogen transmission and test the efficacy of hygiene practices."	ENVO:building	TCAG	Microbial biogeography of public restroom surfaces	NA	0	"Ten surfaces (door handles into and out of the restroom, handles into and out of a restroom stall, faucet handles, soap dispenser, toilet seat, toilet flush handle, floor around the toilet and floor around the sink) in six male and six female restrooms evenly distributed across two buildings on the University of Colorado at Boulder campus were sampled on a single day in November 2010. Surfaces where sampled using sterile, cotton-tipped swabs as described previously?[14],?[15]. As the 12 restrooms were nearly identical in design, we were able to swab the same area at each location between restrooms. In order to characterize tap water communities as a potential source of bacteria, 1 L of faucet water from six of the restrooms (each building having the same water source for each restroom sampled) was collected and filtered through 0.2 ?m bottle top filters (Nalgene, Rochester, NY, USA)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	22132229	NA	NA	NA	Gilberto Flores	CCME	GAZ:United States of America	NA	NA	0	NA	NA	NA	NA	n	NA	Noah Fierer	Flores_restroom_surface_biogeography	pyrosequencing	NA	NA	NA	restroom surface metagenome	3	ENVO:city	Titanium	NA	NA	NA	NA	NA	flores_restroom_surface_biogeography	restroom surface biogeography	256318	CCME	NA	NA	NA	1335	"Genomic DNA was extracted from the swabs and filters using the MO BIO PowerSoil DNA isolation kit following the manufacturer's protocol with the modifications of Fierer?et al.?[14]. A portion of the 16 S rRNA gene spanning the V1_V2 regions was amplified using the primer set (27 F/338R), PCR mixture conditions and thermal cycling conditions described in Fierer?et al.?[15]. PCR amplicons of triplicate reactions for each sample were pooled at approximately equal amounts and pyrosequenced at 454 Life Sciences (Branford, CT, USA) on their GS Junior system. A total of 337,333 high-quality partial 16 S rRNA gene sequences were obtained from 101 of the 120 surface samples collected, averaging approximately 3,340 sequences per sample (ranging from 513_6,771) (Table S1) in 4 GS Junior runs, with the best run containing 116,004 high-quality reads. An additional 16,416 sequences (ranging from 2161_5084 per sample) were generated for five of the six water samples collected for source tracking analysis. Each sample was amplified with a unique barcode to enable multiplexing in the GS Junior runs. The barcoded sequencing reads can be separated by data analysis software providing high confidence in assigning sequencing read to each sample. Sequence data generated as part of this study is available upon request by contacting the corresponding author."	NA	NA	y	NA	NA	ENVO:surface	16S rRNA	Porter	NA	V2	CCME	NA	Toilet flush handle	NA	NA	NA	40.01499	NA	NA	Toilet flush handle
EKAF5.489565	GACATCGGCTAT	YATGCTGCCTCCCGTAGGAGT	jesse.stombaugh@colorado.edu	NA	restroom surface	Toilet_seat	NA	NA	NA	n	12/13/10	restroom surface metagenome	NA	GFL0IUQ	NA	mimarks-survey	NA	NA	NA	NA	0	NA	NA	1624.1	10-Nov	Female	0	NA	NA	NA	NA	NA	NA	NA	NA	NA	NA	"1,swab"	A	NA	Microbial biogeography of public restroom surfaces	NA	NA	-105.27055	y	NA	CCME	NA	NA	NA	"We spend the majority of our lives indoors where we are constantly exposed to bacteria residing on surfaces. However, the diversity of these surface-associated communities is largely unknown. We explored the biogeographical patterns exhibited by bacteria across ten surfaces within each of twelve public restrooms. Using high-throughput barcoded pyrosequencing of the 16S rRNA gene, we identified 19 bacterial phyla across all surfaces. Most sequences belonged to four phyla: Actinobacteria, Bacteriodetes, Firmicutes and Proteobacteria. The communities clustered into three general categories: those found on surfaces associated with toilets, those on the restroom floor, and those found on surfaces routinely touched with hands. On toilet surfaces, gut-associated taxa were more prevalent, suggesting fecal contamination of these surfaces. Floor surfaces were the most diverse of all communities and contained several taxa commonly found in soils. Skin-associated bacteria, especially the Propionibacteriaceae, dominated surfaces routinely touched with our hands. Certain taxa were more common in female than in male restrooms as vagina-associated Lactobacillaceae were widely distributed in female restrooms, likely from urine contamination. Use of the SourceTracker algorithm confirmed many of our taxonomic observations as human skin was the primary source of bacteria on restroom surfaces. Overall, these results demonstrate that restroom surfaces host relatively diverse microbial communities dominated by human-associated bacteria with clear linkages between communities on or in different body sites and those communities found on restroom surfaces. More generally, this work is relevant to the public health field as we show that human associated microbes are commonly found on restroom surfaces suggesting that bacterial pathogens could readily be transmitted between individuals by the touching of surfaces. Furthermore, we demonstrate that we can use high-throughput analyses of bacterial communities to determine sources of bacteria on indoor surfaces, an approach which could be used to track pathogen transmission and test the efficacy of hygiene practices."	ENVO:building	TCAG	Microbial biogeography of public restroom surfaces	NA	0	"Ten surfaces (door handles into and out of the restroom, handles into and out of a restroom stall, faucet handles, soap dispenser, toilet seat, toilet flush handle, floor around the toilet and floor around the sink) in six male and six female restrooms evenly distributed across two buildings on the University of Colorado at Boulder campus were sampled on a single day in November 2010. Surfaces where sampled using sterile, cotton-tipped swabs as described previously?[14],?[15]. As the 12 restrooms were nearly identical in design, we were able to swab the same area at each location between restrooms. In order to characterize tap water communities as a potential source of bacteria, 1 L of faucet water from six of the restrooms (each building having the same water source for each restroom sampled) was collected and filtered through 0.2 ?m bottle top filters (Nalgene, Rochester, NY, USA)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	22132229	NA	NA	NA	Gilberto Flores	CCME	GAZ:United States of America	NA	NA	0	NA	NA	NA	NA	n	NA	Noah Fierer	Flores_restroom_surface_biogeography	pyrosequencing	NA	NA	NA	restroom surface metagenome	1	ENVO:city	Titanium	NA	NA	NA	NA	NA	flores_restroom_surface_biogeography	restroom surface biogeography	256318	CCME	NA	NA	NA	1335	"Genomic DNA was extracted from the swabs and filters using the MO BIO PowerSoil DNA isolation kit following the manufacturer's protocol with the modifications of Fierer?et al.?[14]. A portion of the 16 S rRNA gene spanning the V1_V2 regions was amplified using the primer set (27 F/338R), PCR mixture conditions and thermal cycling conditions described in Fierer?et al.?[15]. PCR amplicons of triplicate reactions for each sample were pooled at approximately equal amounts and pyrosequenced at 454 Life Sciences (Branford, CT, USA) on their GS Junior system. A total of 337,333 high-quality partial 16 S rRNA gene sequences were obtained from 101 of the 120 surface samples collected, averaging approximately 3,340 sequences per sample (ranging from 513_6,771) (Table S1) in 4 GS Junior runs, with the best run containing 116,004 high-quality reads. An additional 16,416 sequences (ranging from 2161_5084 per sample) were generated for five of the six water samples collected for source tracking analysis. Each sample was amplified with a unique barcode to enable multiplexing in the GS Junior runs. The barcoded sequencing reads can be separated by data analysis software providing high confidence in assigning sequencing read to each sample. Sequence data generated as part of this study is available upon request by contacting the corresponding author."	NA	NA	y	NA	NA	ENVO:surface	16S rRNA	Ekeley	NA	V2	CCME	NA	Toilet seat	NA	NA	NA	40.01499	NA	NA	Toilet seat
EKAM5.489530	GAACATGATGAG	YATGCTGCCTCCCGTAGGAGT	jesse.stombaugh@colorado.edu	NA	restroom surface	Toilet_seat	NA	NA	NA	n	12/13/10	restroom surface metagenome	NA	GFL0IUQ	NA	mimarks-survey	NA	NA	NA	NA	0	NA	NA	1624.1	10-Nov	Male	0	NA	NA	NA	NA	NA	NA	NA	NA	NA	NA	"1,swab"	A	NA	Microbial biogeography of public restroom surfaces	NA	NA	-105.27055	y	NA	CCME	NA	NA	NA	"We spend the majority of our lives indoors where we are constantly exposed to bacteria residing on surfaces. However, the diversity of these surface-associated communities is largely unknown. We explored the biogeographical patterns exhibited by bacteria across ten surfaces within each of twelve public restrooms. Using high-throughput barcoded pyrosequencing of the 16S rRNA gene, we identified 19 bacterial phyla across all surfaces. Most sequences belonged to four phyla: Actinobacteria, Bacteriodetes, Firmicutes and Proteobacteria. The communities clustered into three general categories: those found on surfaces associated with toilets, those on the restroom floor, and those found on surfaces routinely touched with hands. On toilet surfaces, gut-associated taxa were more prevalent, suggesting fecal contamination of these surfaces. Floor surfaces were the most diverse of all communities and contained several taxa commonly found in soils. Skin-associated bacteria, especially the Propionibacteriaceae, dominated surfaces routinely touched with our hands. Certain taxa were more common in female than in male restrooms as vagina-associated Lactobacillaceae were widely distributed in female restrooms, likely from urine contamination. Use of the SourceTracker algorithm confirmed many of our taxonomic observations as human skin was the primary source of bacteria on restroom surfaces. Overall, these results demonstrate that restroom surfaces host relatively diverse microbial communities dominated by human-associated bacteria with clear linkages between communities on or in different body sites and those communities found on restroom surfaces. More generally, this work is relevant to the public health field as we show that human associated microbes are commonly found on restroom surfaces suggesting that bacterial pathogens could readily be transmitted between individuals by the touching of surfaces. Furthermore, we demonstrate that we can use high-throughput analyses of bacterial communities to determine sources of bacteria on indoor surfaces, an approach which could be used to track pathogen transmission and test the efficacy of hygiene practices."	ENVO:building	TCAG	Microbial biogeography of public restroom surfaces	NA	0	"Ten surfaces (door handles into and out of the restroom, handles into and out of a restroom stall, faucet handles, soap dispenser, toilet seat, toilet flush handle, floor around the toilet and floor around the sink) in six male and six female restrooms evenly distributed across two buildings on the University of Colorado at Boulder campus were sampled on a single day in November 2010. Surfaces where sampled using sterile, cotton-tipped swabs as described previously?[14],?[15]. As the 12 restrooms were nearly identical in design, we were able to swab the same area at each location between restrooms. In order to characterize tap water communities as a potential source of bacteria, 1 L of faucet water from six of the restrooms (each building having the same water source for each restroom sampled) was collected and filtered through 0.2 ?m bottle top filters (Nalgene, Rochester, NY, USA)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	22132229	NA	NA	NA	Gilberto Flores	CCME	GAZ:United States of America	NA	NA	0	NA	NA	NA	NA	n	NA	Noah Fierer	Flores_restroom_surface_biogeography	pyrosequencing	NA	NA	NA	restroom surface metagenome	1	ENVO:city	Titanium	NA	NA	NA	NA	NA	flores_restroom_surface_biogeography	restroom surface biogeography	256318	CCME	NA	NA	NA	1335	"Genomic DNA was extracted from the swabs and filters using the MO BIO PowerSoil DNA isolation kit following the manufacturer's protocol with the modifications of Fierer?et al.?[14]. A portion of the 16 S rRNA gene spanning the V1_V2 regions was amplified using the primer set (27 F/338R), PCR mixture conditions and thermal cycling conditions described in Fierer?et al.?[15]. PCR amplicons of triplicate reactions for each sample were pooled at approximately equal amounts and pyrosequenced at 454 Life Sciences (Branford, CT, USA) on their GS Junior system. A total of 337,333 high-quality partial 16 S rRNA gene sequences were obtained from 101 of the 120 surface samples collected, averaging approximately 3,340 sequences per sample (ranging from 513_6,771) (Table S1) in 4 GS Junior runs, with the best run containing 116,004 high-quality reads. An additional 16,416 sequences (ranging from 2161_5084 per sample) were generated for five of the six water samples collected for source tracking analysis. Each sample was amplified with a unique barcode to enable multiplexing in the GS Junior runs. The barcoded sequencing reads can be separated by data analysis software providing high confidence in assigning sequencing read to each sample. Sequence data generated as part of this study is available upon request by contacting the corresponding author."	NA	NA	y	NA	NA	ENVO:surface	16S rRNA	Ekeley	NA	V2	CCME	NA	Toilet seat	NA	NA	NA	40.01499	NA	NA	Toilet seat
EKBF5.489551	GAGATGCCGACT	YATGCTGCCTCCCGTAGGAGT	jesse.stombaugh@colorado.edu	NA	restroom surface	Toilet_seat	NA	NA	NA	n	12/13/10	restroom surface metagenome	NA	GFL0IUQ	NA	mimarks-survey	NA	NA	NA	NA	0	NA	NA	1624.1	10-Nov	Female	0	NA	NA	NA	NA	NA	NA	NA	NA	NA	NA	"1,swab"	B	NA	Microbial biogeography of public restroom surfaces	NA	NA	-105.27055	y	NA	CCME	NA	NA	NA	"We spend the majority of our lives indoors where we are constantly exposed to bacteria residing on surfaces. However, the diversity of these surface-associated communities is largely unknown. We explored the biogeographical patterns exhibited by bacteria across ten surfaces within each of twelve public restrooms. Using high-throughput barcoded pyrosequencing of the 16S rRNA gene, we identified 19 bacterial phyla across all surfaces. Most sequences belonged to four phyla: Actinobacteria, Bacteriodetes, Firmicutes and Proteobacteria. The communities clustered into three general categories: those found on surfaces associated with toilets, those on the restroom floor, and those found on surfaces routinely touched with hands. On toilet surfaces, gut-associated taxa were more prevalent, suggesting fecal contamination of these surfaces. Floor surfaces were the most diverse of all communities and contained several taxa commonly found in soils. Skin-associated bacteria, especially the Propionibacteriaceae, dominated surfaces routinely touched with our hands. Certain taxa were more common in female than in male restrooms as vagina-associated Lactobacillaceae were widely distributed in female restrooms, likely from urine contamination. Use of the SourceTracker algorithm confirmed many of our taxonomic observations as human skin was the primary source of bacteria on restroom surfaces. Overall, these results demonstrate that restroom surfaces host relatively diverse microbial communities dominated by human-associated bacteria with clear linkages between communities on or in different body sites and those communities found on restroom surfaces. More generally, this work is relevant to the public health field as we show that human associated microbes are commonly found on restroom surfaces suggesting that bacterial pathogens could readily be transmitted between individuals by the touching of surfaces. Furthermore, we demonstrate that we can use high-throughput analyses of bacterial communities to determine sources of bacteria on indoor surfaces, an approach which could be used to track pathogen transmission and test the efficacy of hygiene practices."	ENVO:building	TCAG	Microbial biogeography of public restroom surfaces	NA	0	"Ten surfaces (door handles into and out of the restroom, handles into and out of a restroom stall, faucet handles, soap dispenser, toilet seat, toilet flush handle, floor around the toilet and floor around the sink) in six male and six female restrooms evenly distributed across two buildings on the University of Colorado at Boulder campus were sampled on a single day in November 2010. Surfaces where sampled using sterile, cotton-tipped swabs as described previously?[14],?[15]. As the 12 restrooms were nearly identical in design, we were able to swab the same area at each location between restrooms. In order to characterize tap water communities as a potential source of bacteria, 1 L of faucet water from six of the restrooms (each building having the same water source for each restroom sampled) was collected and filtered through 0.2 ?m bottle top filters (Nalgene, Rochester, NY, USA)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	22132229	NA	NA	NA	Gilberto Flores	CCME	GAZ:United States of America	NA	NA	0	NA	NA	NA	NA	n	NA	Noah Fierer	Flores_restroom_surface_biogeography	pyrosequencing	NA	NA	NA	restroom surface metagenome	2	ENVO:city	Titanium	NA	NA	NA	NA	NA	flores_restroom_surface_biogeography	restroom surface biogeography	256318	CCME	NA	NA	NA	1335	"Genomic DNA was extracted from the swabs and filters using the MO BIO PowerSoil DNA isolation kit following the manufacturer's protocol with the modifications of Fierer?et al.?[14]. A portion of the 16 S rRNA gene spanning the V1_V2 regions was amplified using the primer set (27 F/338R), PCR mixture conditions and thermal cycling conditions described in Fierer?et al.?[15]. PCR amplicons of triplicate reactions for each sample were pooled at approximately equal amounts and pyrosequenced at 454 Life Sciences (Branford, CT, USA) on their GS Junior system. A total of 337,333 high-quality partial 16 S rRNA gene sequences were obtained from 101 of the 120 surface samples collected, averaging approximately 3,340 sequences per sample (ranging from 513_6,771) (Table S1) in 4 GS Junior runs, with the best run containing 116,004 high-quality reads. An additional 16,416 sequences (ranging from 2161_5084 per sample) were generated for five of the six water samples collected for source tracking analysis. Each sample was amplified with a unique barcode to enable multiplexing in the GS Junior runs. The barcoded sequencing reads can be separated by data analysis software providing high confidence in assigning sequencing read to each sample. Sequence data generated as part of this study is available upon request by contacting the corresponding author."	NA	NA	y	NA	NA	ENVO:surface	16S rRNA	Ekeley	NA	V2	CCME	NA	Toilet seat	NA	NA	NA	40.01499	NA	NA	Toilet seat
EKBM5.489472	GACTCACTCAAT	YATGCTGCCTCCCGTAGGAGT	jesse.stombaugh@colorado.edu	NA	restroom surface	Toilet_seat	NA	NA	NA	n	12/13/10	restroom surface metagenome	NA	GFL0IUQ	NA	mimarks-survey	NA	NA	NA	NA	0	NA	NA	1624.1	10-Nov	Male	0	NA	NA	NA	NA	NA	NA	NA	NA	NA	NA	"1,swab"	B	NA	Microbial biogeography of public restroom surfaces	NA	NA	-105.27055	y	NA	CCME	NA	NA	NA	"We spend the majority of our lives indoors where we are constantly exposed to bacteria residing on surfaces. However, the diversity of these surface-associated communities is largely unknown. We explored the biogeographical patterns exhibited by bacteria across ten surfaces within each of twelve public restrooms. Using high-throughput barcoded pyrosequencing of the 16S rRNA gene, we identified 19 bacterial phyla across all surfaces. Most sequences belonged to four phyla: Actinobacteria, Bacteriodetes, Firmicutes and Proteobacteria. The communities clustered into three general categories: those found on surfaces associated with toilets, those on the restroom floor, and those found on surfaces routinely touched with hands. On toilet surfaces, gut-associated taxa were more prevalent, suggesting fecal contamination of these surfaces. Floor surfaces were the most diverse of all communities and contained several taxa commonly found in soils. Skin-associated bacteria, especially the Propionibacteriaceae, dominated surfaces routinely touched with our hands. Certain taxa were more common in female than in male restrooms as vagina-associated Lactobacillaceae were widely distributed in female restrooms, likely from urine contamination. Use of the SourceTracker algorithm confirmed many of our taxonomic observations as human skin was the primary source of bacteria on restroom surfaces. Overall, these results demonstrate that restroom surfaces host relatively diverse microbial communities dominated by human-associated bacteria with clear linkages between communities on or in different body sites and those communities found on restroom surfaces. More generally, this work is relevant to the public health field as we show that human associated microbes are commonly found on restroom surfaces suggesting that bacterial pathogens could readily be transmitted between individuals by the touching of surfaces. Furthermore, we demonstrate that we can use high-throughput analyses of bacterial communities to determine sources of bacteria on indoor surfaces, an approach which could be used to track pathogen transmission and test the efficacy of hygiene practices."	ENVO:building	TCAG	Microbial biogeography of public restroom surfaces	NA	0	"Ten surfaces (door handles into and out of the restroom, handles into and out of a restroom stall, faucet handles, soap dispenser, toilet seat, toilet flush handle, floor around the toilet and floor around the sink) in six male and six female restrooms evenly distributed across two buildings on the University of Colorado at Boulder campus were sampled on a single day in November 2010. Surfaces where sampled using sterile, cotton-tipped swabs as described previously?[14],?[15]. As the 12 restrooms were nearly identical in design, we were able to swab the same area at each location between restrooms. In order to characterize tap water communities as a potential source of bacteria, 1 L of faucet water from six of the restrooms (each building having the same water source for each restroom sampled) was collected and filtered through 0.2 ?m bottle top filters (Nalgene, Rochester, NY, USA)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	22132229	NA	NA	NA	Gilberto Flores	CCME	GAZ:United States of America	NA	NA	0	NA	NA	NA	NA	n	NA	Noah Fierer	Flores_restroom_surface_biogeography	pyrosequencing	NA	NA	NA	restroom surface metagenome	2	ENVO:city	Titanium	NA	NA	NA	NA	NA	flores_restroom_surface_biogeography	restroom surface biogeography	256318	CCME	NA	NA	NA	1335	"Genomic DNA was extracted from the swabs and filters using the MO BIO PowerSoil DNA isolation kit following the manufacturer's protocol with the modifications of Fierer?et al.?[14]. A portion of the 16 S rRNA gene spanning the V1_V2 regions was amplified using the primer set (27 F/338R), PCR mixture conditions and thermal cycling conditions described in Fierer?et al.?[15]. PCR amplicons of triplicate reactions for each sample were pooled at approximately equal amounts and pyrosequenced at 454 Life Sciences (Branford, CT, USA) on their GS Junior system. A total of 337,333 high-quality partial 16 S rRNA gene sequences were obtained from 101 of the 120 surface samples collected, averaging approximately 3,340 sequences per sample (ranging from 513_6,771) (Table S1) in 4 GS Junior runs, with the best run containing 116,004 high-quality reads. An additional 16,416 sequences (ranging from 2161_5084 per sample) were generated for five of the six water samples collected for source tracking analysis. Each sample was amplified with a unique barcode to enable multiplexing in the GS Junior runs. The barcoded sequencing reads can be separated by data analysis software providing high confidence in assigning sequencing read to each sample. Sequence data generated as part of this study is available upon request by contacting the corresponding author."	NA	NA	y	NA	NA	ENVO:surface	16S rRNA	Ekeley	NA	V2	CCME	NA	Toilet seat	NA	NA	NA	40.01499	NA	NA	Toilet seat
EKCF5.489488	GATCTTCAGTAC	YATGCTGCCTCCCGTAGGAGT	jesse.stombaugh@colorado.edu	NA	restroom surface	Toilet_seat	NA	NA	NA	n	12/13/10	restroom surface metagenome	NA	GFL0IUQ	NA	mimarks-survey	NA	NA	NA	NA	0	NA	NA	1624.1	10-Nov	Female	0	NA	NA	NA	NA	NA	NA	NA	NA	NA	NA	"1,swab"	C	NA	Microbial biogeography of public restroom surfaces	NA	NA	-105.27055	y	NA	CCME	NA	NA	NA	"We spend the majority of our lives indoors where we are constantly exposed to bacteria residing on surfaces. However, the diversity of these surface-associated communities is largely unknown. We explored the biogeographical patterns exhibited by bacteria across ten surfaces within each of twelve public restrooms. Using high-throughput barcoded pyrosequencing of the 16S rRNA gene, we identified 19 bacterial phyla across all surfaces. Most sequences belonged to four phyla: Actinobacteria, Bacteriodetes, Firmicutes and Proteobacteria. The communities clustered into three general categories: those found on surfaces associated with toilets, those on the restroom floor, and those found on surfaces routinely touched with hands. On toilet surfaces, gut-associated taxa were more prevalent, suggesting fecal contamination of these surfaces. Floor surfaces were the most diverse of all communities and contained several taxa commonly found in soils. Skin-associated bacteria, especially the Propionibacteriaceae, dominated surfaces routinely touched with our hands. Certain taxa were more common in female than in male restrooms as vagina-associated Lactobacillaceae were widely distributed in female restrooms, likely from urine contamination. Use of the SourceTracker algorithm confirmed many of our taxonomic observations as human skin was the primary source of bacteria on restroom surfaces. Overall, these results demonstrate that restroom surfaces host relatively diverse microbial communities dominated by human-associated bacteria with clear linkages between communities on or in different body sites and those communities found on restroom surfaces. More generally, this work is relevant to the public health field as we show that human associated microbes are commonly found on restroom surfaces suggesting that bacterial pathogens could readily be transmitted between individuals by the touching of surfaces. Furthermore, we demonstrate that we can use high-throughput analyses of bacterial communities to determine sources of bacteria on indoor surfaces, an approach which could be used to track pathogen transmission and test the efficacy of hygiene practices."	ENVO:building	TCAG	Microbial biogeography of public restroom surfaces	NA	0	"Ten surfaces (door handles into and out of the restroom, handles into and out of a restroom stall, faucet handles, soap dispenser, toilet seat, toilet flush handle, floor around the toilet and floor around the sink) in six male and six female restrooms evenly distributed across two buildings on the University of Colorado at Boulder campus were sampled on a single day in November 2010. Surfaces where sampled using sterile, cotton-tipped swabs as described previously?[14],?[15]. As the 12 restrooms were nearly identical in design, we were able to swab the same area at each location between restrooms. In order to characterize tap water communities as a potential source of bacteria, 1 L of faucet water from six of the restrooms (each building having the same water source for each restroom sampled) was collected and filtered through 0.2 ?m bottle top filters (Nalgene, Rochester, NY, USA)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	22132229	NA	NA	NA	Gilberto Flores	CCME	GAZ:United States of America	NA	NA	0	NA	NA	NA	NA	n	NA	Noah Fierer	Flores_restroom_surface_biogeography	pyrosequencing	NA	NA	NA	restroom surface metagenome	3	ENVO:city	Titanium	NA	NA	NA	NA	NA	flores_restroom_surface_biogeography	restroom surface biogeography	256318	CCME	NA	NA	NA	1335	"Genomic DNA was extracted from the swabs and filters using the MO BIO PowerSoil DNA isolation kit following the manufacturer's protocol with the modifications of Fierer?et al.?[14]. A portion of the 16 S rRNA gene spanning the V1_V2 regions was amplified using the primer set (27 F/338R), PCR mixture conditions and thermal cycling conditions described in Fierer?et al.?[15]. PCR amplicons of triplicate reactions for each sample were pooled at approximately equal amounts and pyrosequenced at 454 Life Sciences (Branford, CT, USA) on their GS Junior system. A total of 337,333 high-quality partial 16 S rRNA gene sequences were obtained from 101 of the 120 surface samples collected, averaging approximately 3,340 sequences per sample (ranging from 513_6,771) (Table S1) in 4 GS Junior runs, with the best run containing 116,004 high-quality reads. An additional 16,416 sequences (ranging from 2161_5084 per sample) were generated for five of the six water samples collected for source tracking analysis. Each sample was amplified with a unique barcode to enable multiplexing in the GS Junior runs. The barcoded sequencing reads can be separated by data analysis software providing high confidence in assigning sequencing read to each sample. Sequence data generated as part of this study is available upon request by contacting the corresponding author."	NA	NA	y	NA	NA	ENVO:surface	16S rRNA	Ekeley	NA	V2	CCME	NA	Toilet seat	NA	NA	NA	40.01499	NA	NA	Toilet seat
EKCM5.489561	GATACGTCCTGA	YATGCTGCCTCCCGTAGGAGT	jesse.stombaugh@colorado.edu	NA	restroom surface	Toilet_seat	NA	NA	NA	n	12/13/10	restroom surface metagenome	NA	GFL0IUQ	NA	mimarks-survey	NA	NA	NA	NA	0	NA	NA	1624.1	10-Nov	Male	0	NA	NA	NA	NA	NA	NA	NA	NA	NA	NA	"1,swab"	C	NA	Microbial biogeography of public restroom surfaces	NA	NA	-105.27055	y	NA	CCME	NA	NA	NA	"We spend the majority of our lives indoors where we are constantly exposed to bacteria residing on surfaces. However, the diversity of these surface-associated communities is largely unknown. We explored the biogeographical patterns exhibited by bacteria across ten surfaces within each of twelve public restrooms. Using high-throughput barcoded pyrosequencing of the 16S rRNA gene, we identified 19 bacterial phyla across all surfaces. Most sequences belonged to four phyla: Actinobacteria, Bacteriodetes, Firmicutes and Proteobacteria. The communities clustered into three general categories: those found on surfaces associated with toilets, those on the restroom floor, and those found on surfaces routinely touched with hands. On toilet surfaces, gut-associated taxa were more prevalent, suggesting fecal contamination of these surfaces. Floor surfaces were the most diverse of all communities and contained several taxa commonly found in soils. Skin-associated bacteria, especially the Propionibacteriaceae, dominated surfaces routinely touched with our hands. Certain taxa were more common in female than in male restrooms as vagina-associated Lactobacillaceae were widely distributed in female restrooms, likely from urine contamination. Use of the SourceTracker algorithm confirmed many of our taxonomic observations as human skin was the primary source of bacteria on restroom surfaces. Overall, these results demonstrate that restroom surfaces host relatively diverse microbial communities dominated by human-associated bacteria with clear linkages between communities on or in different body sites and those communities found on restroom surfaces. More generally, this work is relevant to the public health field as we show that human associated microbes are commonly found on restroom surfaces suggesting that bacterial pathogens could readily be transmitted between individuals by the touching of surfaces. Furthermore, we demonstrate that we can use high-throughput analyses of bacterial communities to determine sources of bacteria on indoor surfaces, an approach which could be used to track pathogen transmission and test the efficacy of hygiene practices."	ENVO:building	TCAG	Microbial biogeography of public restroom surfaces	NA	0	"Ten surfaces (door handles into and out of the restroom, handles into and out of a restroom stall, faucet handles, soap dispenser, toilet seat, toilet flush handle, floor around the toilet and floor around the sink) in six male and six female restrooms evenly distributed across two buildings on the University of Colorado at Boulder campus were sampled on a single day in November 2010. Surfaces where sampled using sterile, cotton-tipped swabs as described previously?[14],?[15]. As the 12 restrooms were nearly identical in design, we were able to swab the same area at each location between restrooms. In order to characterize tap water communities as a potential source of bacteria, 1 L of faucet water from six of the restrooms (each building having the same water source for each restroom sampled) was collected and filtered through 0.2 ?m bottle top filters (Nalgene, Rochester, NY, USA)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	22132229	NA	NA	NA	Gilberto Flores	CCME	GAZ:United States of America	NA	NA	0	NA	NA	NA	NA	n	NA	Noah Fierer	Flores_restroom_surface_biogeography	pyrosequencing	NA	NA	NA	restroom surface metagenome	3	ENVO:city	Titanium	NA	NA	NA	NA	NA	flores_restroom_surface_biogeography	restroom surface biogeography	256318	CCME	NA	NA	NA	1335	"Genomic DNA was extracted from the swabs and filters using the MO BIO PowerSoil DNA isolation kit following the manufacturer's protocol with the modifications of Fierer?et al.?[14]. A portion of the 16 S rRNA gene spanning the V1_V2 regions was amplified using the primer set (27 F/338R), PCR mixture conditions and thermal cycling conditions described in Fierer?et al.?[15]. PCR amplicons of triplicate reactions for each sample were pooled at approximately equal amounts and pyrosequenced at 454 Life Sciences (Branford, CT, USA) on their GS Junior system. A total of 337,333 high-quality partial 16 S rRNA gene sequences were obtained from 101 of the 120 surface samples collected, averaging approximately 3,340 sequences per sample (ranging from 513_6,771) (Table S1) in 4 GS Junior runs, with the best run containing 116,004 high-quality reads. An additional 16,416 sequences (ranging from 2161_5084 per sample) were generated for five of the six water samples collected for source tracking analysis. Each sample was amplified with a unique barcode to enable multiplexing in the GS Junior runs. The barcoded sequencing reads can be separated by data analysis software providing high confidence in assigning sequencing read to each sample. Sequence data generated as part of this study is available upon request by contacting the corresponding author."	NA	NA	y	NA	NA	ENVO:surface	16S rRNA	Ekeley	NA	V2	CCME	NA	Toilet seat	NA	NA	NA	40.01499	NA	NA	Toilet seat
PTAF5.489560	GCATGTGCATGT	YATGCTGCCTCCCGTAGGAGT	jesse.stombaugh@colorado.edu	NA	restroom surface	Toilet_seat	NA	NA	NA	n	12/13/10	restroom surface metagenome	NA	GFL0IUQ	NA	mimarks-survey	NA	NA	NA	NA	0	NA	NA	1624.1	10-Nov	Female	0	NA	NA	NA	NA	NA	NA	NA	NA	NA	NA	"1,swab"	A	NA	Microbial biogeography of public restroom surfaces	NA	NA	-105.27055	y	NA	CCME	NA	NA	NA	"We spend the majority of our lives indoors where we are constantly exposed to bacteria residing on surfaces. However, the diversity of these surface-associated communities is largely unknown. We explored the biogeographical patterns exhibited by bacteria across ten surfaces within each of twelve public restrooms. Using high-throughput barcoded pyrosequencing of the 16S rRNA gene, we identified 19 bacterial phyla across all surfaces. Most sequences belonged to four phyla: Actinobacteria, Bacteriodetes, Firmicutes and Proteobacteria. The communities clustered into three general categories: those found on surfaces associated with toilets, those on the restroom floor, and those found on surfaces routinely touched with hands. On toilet surfaces, gut-associated taxa were more prevalent, suggesting fecal contamination of these surfaces. Floor surfaces were the most diverse of all communities and contained several taxa commonly found in soils. Skin-associated bacteria, especially the Propionibacteriaceae, dominated surfaces routinely touched with our hands. Certain taxa were more common in female than in male restrooms as vagina-associated Lactobacillaceae were widely distributed in female restrooms, likely from urine contamination. Use of the SourceTracker algorithm confirmed many of our taxonomic observations as human skin was the primary source of bacteria on restroom surfaces. Overall, these results demonstrate that restroom surfaces host relatively diverse microbial communities dominated by human-associated bacteria with clear linkages between communities on or in different body sites and those communities found on restroom surfaces. More generally, this work is relevant to the public health field as we show that human associated microbes are commonly found on restroom surfaces suggesting that bacterial pathogens could readily be transmitted between individuals by the touching of surfaces. Furthermore, we demonstrate that we can use high-throughput analyses of bacterial communities to determine sources of bacteria on indoor surfaces, an approach which could be used to track pathogen transmission and test the efficacy of hygiene practices."	ENVO:building	TCAG	Microbial biogeography of public restroom surfaces	NA	0	"Ten surfaces (door handles into and out of the restroom, handles into and out of a restroom stall, faucet handles, soap dispenser, toilet seat, toilet flush handle, floor around the toilet and floor around the sink) in six male and six female restrooms evenly distributed across two buildings on the University of Colorado at Boulder campus were sampled on a single day in November 2010. Surfaces where sampled using sterile, cotton-tipped swabs as described previously?[14],?[15]. As the 12 restrooms were nearly identical in design, we were able to swab the same area at each location between restrooms. In order to characterize tap water communities as a potential source of bacteria, 1 L of faucet water from six of the restrooms (each building having the same water source for each restroom sampled) was collected and filtered through 0.2 ?m bottle top filters (Nalgene, Rochester, NY, USA)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	22132229	NA	NA	NA	Gilberto Flores	CCME	GAZ:United States of America	NA	NA	0	NA	NA	NA	NA	n	NA	Noah Fierer	Flores_restroom_surface_biogeography	pyrosequencing	NA	NA	NA	restroom surface metagenome	1	ENVO:city	Titanium	NA	NA	NA	NA	NA	flores_restroom_surface_biogeography	restroom surface biogeography	256318	CCME	NA	NA	NA	1335	"Genomic DNA was extracted from the swabs and filters using the MO BIO PowerSoil DNA isolation kit following the manufacturer's protocol with the modifications of Fierer?et al.?[14]. A portion of the 16 S rRNA gene spanning the V1_V2 regions was amplified using the primer set (27 F/338R), PCR mixture conditions and thermal cycling conditions described in Fierer?et al.?[15]. PCR amplicons of triplicate reactions for each sample were pooled at approximately equal amounts and pyrosequenced at 454 Life Sciences (Branford, CT, USA) on their GS Junior system. A total of 337,333 high-quality partial 16 S rRNA gene sequences were obtained from 101 of the 120 surface samples collected, averaging approximately 3,340 sequences per sample (ranging from 513_6,771) (Table S1) in 4 GS Junior runs, with the best run containing 116,004 high-quality reads. An additional 16,416 sequences (ranging from 2161_5084 per sample) were generated for five of the six water samples collected for source tracking analysis. Each sample was amplified with a unique barcode to enable multiplexing in the GS Junior runs. The barcoded sequencing reads can be separated by data analysis software providing high confidence in assigning sequencing read to each sample. Sequence data generated as part of this study is available upon request by contacting the corresponding author."	NA	NA	y	NA	NA	ENVO:surface	16S rRNA	Porter	NA	V2	CCME	NA	Toilet seat	NA	NA	NA	40.01499	NA	NA	Toilet seat
PTAM5.489496	GCACTGAGACGT	YATGCTGCCTCCCGTAGGAGT	jesse.stombaugh@colorado.edu	NA	restroom surface	Toilet_seat	NA	NA	NA	n	12/13/10	restroom surface metagenome	NA	GFL0IUQ	NA	mimarks-survey	NA	NA	NA	NA	0	NA	NA	1624.1	10-Nov	Male	0	NA	NA	NA	NA	NA	NA	NA	NA	NA	NA	"1,swab"	A	NA	Microbial biogeography of public restroom surfaces	NA	NA	-105.27055	y	NA	CCME	NA	NA	NA	"We spend the majority of our lives indoors where we are constantly exposed to bacteria residing on surfaces. However, the diversity of these surface-associated communities is largely unknown. We explored the biogeographical patterns exhibited by bacteria across ten surfaces within each of twelve public restrooms. Using high-throughput barcoded pyrosequencing of the 16S rRNA gene, we identified 19 bacterial phyla across all surfaces. Most sequences belonged to four phyla: Actinobacteria, Bacteriodetes, Firmicutes and Proteobacteria. The communities clustered into three general categories: those found on surfaces associated with toilets, those on the restroom floor, and those found on surfaces routinely touched with hands. On toilet surfaces, gut-associated taxa were more prevalent, suggesting fecal contamination of these surfaces. Floor surfaces were the most diverse of all communities and contained several taxa commonly found in soils. Skin-associated bacteria, especially the Propionibacteriaceae, dominated surfaces routinely touched with our hands. Certain taxa were more common in female than in male restrooms as vagina-associated Lactobacillaceae were widely distributed in female restrooms, likely from urine contamination. Use of the SourceTracker algorithm confirmed many of our taxonomic observations as human skin was the primary source of bacteria on restroom surfaces. Overall, these results demonstrate that restroom surfaces host relatively diverse microbial communities dominated by human-associated bacteria with clear linkages between communities on or in different body sites and those communities found on restroom surfaces. More generally, this work is relevant to the public health field as we show that human associated microbes are commonly found on restroom surfaces suggesting that bacterial pathogens could readily be transmitted between individuals by the touching of surfaces. Furthermore, we demonstrate that we can use high-throughput analyses of bacterial communities to determine sources of bacteria on indoor surfaces, an approach which could be used to track pathogen transmission and test the efficacy of hygiene practices."	ENVO:building	TCAG	Microbial biogeography of public restroom surfaces	NA	0	"Ten surfaces (door handles into and out of the restroom, handles into and out of a restroom stall, faucet handles, soap dispenser, toilet seat, toilet flush handle, floor around the toilet and floor around the sink) in six male and six female restrooms evenly distributed across two buildings on the University of Colorado at Boulder campus were sampled on a single day in November 2010. Surfaces where sampled using sterile, cotton-tipped swabs as described previously?[14],?[15]. As the 12 restrooms were nearly identical in design, we were able to swab the same area at each location between restrooms. In order to characterize tap water communities as a potential source of bacteria, 1 L of faucet water from six of the restrooms (each building having the same water source for each restroom sampled) was collected and filtered through 0.2 ?m bottle top filters (Nalgene, Rochester, NY, USA)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	22132229	NA	NA	NA	Gilberto Flores	CCME	GAZ:United States of America	NA	NA	0	NA	NA	NA	NA	n	NA	Noah Fierer	Flores_restroom_surface_biogeography	pyrosequencing	NA	NA	NA	restroom surface metagenome	1	ENVO:city	Titanium	NA	NA	NA	NA	NA	flores_restroom_surface_biogeography	restroom surface biogeography	256318	CCME	NA	NA	NA	1335	"Genomic DNA was extracted from the swabs and filters using the MO BIO PowerSoil DNA isolation kit following the manufacturer's protocol with the modifications of Fierer?et al.?[14]. A portion of the 16 S rRNA gene spanning the V1_V2 regions was amplified using the primer set (27 F/338R), PCR mixture conditions and thermal cycling conditions described in Fierer?et al.?[15]. PCR amplicons of triplicate reactions for each sample were pooled at approximately equal amounts and pyrosequenced at 454 Life Sciences (Branford, CT, USA) on their GS Junior system. A total of 337,333 high-quality partial 16 S rRNA gene sequences were obtained from 101 of the 120 surface samples collected, averaging approximately 3,340 sequences per sample (ranging from 513_6,771) (Table S1) in 4 GS Junior runs, with the best run containing 116,004 high-quality reads. An additional 16,416 sequences (ranging from 2161_5084 per sample) were generated for five of the six water samples collected for source tracking analysis. Each sample was amplified with a unique barcode to enable multiplexing in the GS Junior runs. The barcoded sequencing reads can be separated by data analysis software providing high confidence in assigning sequencing read to each sample. Sequence data generated as part of this study is available upon request by contacting the corresponding author."	NA	NA	y	NA	NA	ENVO:surface	16S rRNA	Porter	NA	V2	CCME	NA	Toilet seat	NA	NA	NA	40.01499	NA	NA	Toilet seat
PTBF5.489479	GCTGCTGCAATA	YATGCTGCCTCCCGTAGGAGT	jesse.stombaugh@colorado.edu	NA	restroom surface	Toilet_seat	NA	NA	NA	n	12/13/10	restroom surface metagenome	NA	GFL0IUQ	NA	mimarks-survey	NA	NA	NA	NA	0	NA	NA	1624.1	10-Nov	Female	0	NA	NA	NA	NA	NA	NA	NA	NA	NA	NA	"1,swab"	B	NA	Microbial biogeography of public restroom surfaces	NA	NA	-105.27055	y	NA	CCME	NA	NA	NA	"We spend the majority of our lives indoors where we are constantly exposed to bacteria residing on surfaces. However, the diversity of these surface-associated communities is largely unknown. We explored the biogeographical patterns exhibited by bacteria across ten surfaces within each of twelve public restrooms. Using high-throughput barcoded pyrosequencing of the 16S rRNA gene, we identified 19 bacterial phyla across all surfaces. Most sequences belonged to four phyla: Actinobacteria, Bacteriodetes, Firmicutes and Proteobacteria. The communities clustered into three general categories: those found on surfaces associated with toilets, those on the restroom floor, and those found on surfaces routinely touched with hands. On toilet surfaces, gut-associated taxa were more prevalent, suggesting fecal contamination of these surfaces. Floor surfaces were the most diverse of all communities and contained several taxa commonly found in soils. Skin-associated bacteria, especially the Propionibacteriaceae, dominated surfaces routinely touched with our hands. Certain taxa were more common in female than in male restrooms as vagina-associated Lactobacillaceae were widely distributed in female restrooms, likely from urine contamination. Use of the SourceTracker algorithm confirmed many of our taxonomic observations as human skin was the primary source of bacteria on restroom surfaces. Overall, these results demonstrate that restroom surfaces host relatively diverse microbial communities dominated by human-associated bacteria with clear linkages between communities on or in different body sites and those communities found on restroom surfaces. More generally, this work is relevant to the public health field as we show that human associated microbes are commonly found on restroom surfaces suggesting that bacterial pathogens could readily be transmitted between individuals by the touching of surfaces. Furthermore, we demonstrate that we can use high-throughput analyses of bacterial communities to determine sources of bacteria on indoor surfaces, an approach which could be used to track pathogen transmission and test the efficacy of hygiene practices."	ENVO:building	TCAG	Microbial biogeography of public restroom surfaces	NA	0	"Ten surfaces (door handles into and out of the restroom, handles into and out of a restroom stall, faucet handles, soap dispenser, toilet seat, toilet flush handle, floor around the toilet and floor around the sink) in six male and six female restrooms evenly distributed across two buildings on the University of Colorado at Boulder campus were sampled on a single day in November 2010. Surfaces where sampled using sterile, cotton-tipped swabs as described previously?[14],?[15]. As the 12 restrooms were nearly identical in design, we were able to swab the same area at each location between restrooms. In order to characterize tap water communities as a potential source of bacteria, 1 L of faucet water from six of the restrooms (each building having the same water source for each restroom sampled) was collected and filtered through 0.2 ?m bottle top filters (Nalgene, Rochester, NY, USA)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	22132229	NA	NA	NA	Gilberto Flores	CCME	GAZ:United States of America	NA	NA	0	NA	NA	NA	NA	n	NA	Noah Fierer	Flores_restroom_surface_biogeography	pyrosequencing	NA	NA	NA	restroom surface metagenome	2	ENVO:city	Titanium	NA	NA	NA	NA	NA	flores_restroom_surface_biogeography	restroom surface biogeography	256318	CCME	NA	NA	NA	1335	"Genomic DNA was extracted from the swabs and filters using the MO BIO PowerSoil DNA isolation kit following the manufacturer's protocol with the modifications of Fierer?et al.?[14]. A portion of the 16 S rRNA gene spanning the V1_V2 regions was amplified using the primer set (27 F/338R), PCR mixture conditions and thermal cycling conditions described in Fierer?et al.?[15]. PCR amplicons of triplicate reactions for each sample were pooled at approximately equal amounts and pyrosequenced at 454 Life Sciences (Branford, CT, USA) on their GS Junior system. A total of 337,333 high-quality partial 16 S rRNA gene sequences were obtained from 101 of the 120 surface samples collected, averaging approximately 3,340 sequences per sample (ranging from 513_6,771) (Table S1) in 4 GS Junior runs, with the best run containing 116,004 high-quality reads. An additional 16,416 sequences (ranging from 2161_5084 per sample) were generated for five of the six water samples collected for source tracking analysis. Each sample was amplified with a unique barcode to enable multiplexing in the GS Junior runs. The barcoded sequencing reads can be separated by data analysis software providing high confidence in assigning sequencing read to each sample. Sequence data generated as part of this study is available upon request by contacting the corresponding author."	NA	NA	y	NA	NA	ENVO:surface	16S rRNA	Porter	NA	V2	CCME	NA	Toilet seat	NA	NA	NA	40.01499	NA	NA	Toilet seat
PTCF5.489507	ATCACTAGTCAC	YATGCTGCCTCCCGTAGGAGT	jesse.stombaugh@colorado.edu	NA	restroom surface	Toilet_seat	NA	NA	NA	n	12/13/10	restroom surface metagenome	NA	GFL0IUQ	NA	mimarks-survey	NA	NA	NA	NA	0	NA	NA	1624.1	10-Nov	Female	0	NA	NA	NA	NA	NA	NA	NA	NA	NA	NA	"1,swab"	C	NA	Microbial biogeography of public restroom surfaces	NA	NA	-105.27055	y	NA	CCME	NA	NA	NA	"We spend the majority of our lives indoors where we are constantly exposed to bacteria residing on surfaces. However, the diversity of these surface-associated communities is largely unknown. We explored the biogeographical patterns exhibited by bacteria across ten surfaces within each of twelve public restrooms. Using high-throughput barcoded pyrosequencing of the 16S rRNA gene, we identified 19 bacterial phyla across all surfaces. Most sequences belonged to four phyla: Actinobacteria, Bacteriodetes, Firmicutes and Proteobacteria. The communities clustered into three general categories: those found on surfaces associated with toilets, those on the restroom floor, and those found on surfaces routinely touched with hands. On toilet surfaces, gut-associated taxa were more prevalent, suggesting fecal contamination of these surfaces. Floor surfaces were the most diverse of all communities and contained several taxa commonly found in soils. Skin-associated bacteria, especially the Propionibacteriaceae, dominated surfaces routinely touched with our hands. Certain taxa were more common in female than in male restrooms as vagina-associated Lactobacillaceae were widely distributed in female restrooms, likely from urine contamination. Use of the SourceTracker algorithm confirmed many of our taxonomic observations as human skin was the primary source of bacteria on restroom surfaces. Overall, these results demonstrate that restroom surfaces host relatively diverse microbial communities dominated by human-associated bacteria with clear linkages between communities on or in different body sites and those communities found on restroom surfaces. More generally, this work is relevant to the public health field as we show that human associated microbes are commonly found on restroom surfaces suggesting that bacterial pathogens could readily be transmitted between individuals by the touching of surfaces. Furthermore, we demonstrate that we can use high-throughput analyses of bacterial communities to determine sources of bacteria on indoor surfaces, an approach which could be used to track pathogen transmission and test the efficacy of hygiene practices."	ENVO:building	TCAG	Microbial biogeography of public restroom surfaces	NA	0	"Ten surfaces (door handles into and out of the restroom, handles into and out of a restroom stall, faucet handles, soap dispenser, toilet seat, toilet flush handle, floor around the toilet and floor around the sink) in six male and six female restrooms evenly distributed across two buildings on the University of Colorado at Boulder campus were sampled on a single day in November 2010. Surfaces where sampled using sterile, cotton-tipped swabs as described previously?[14],?[15]. As the 12 restrooms were nearly identical in design, we were able to swab the same area at each location between restrooms. In order to characterize tap water communities as a potential source of bacteria, 1 L of faucet water from six of the restrooms (each building having the same water source for each restroom sampled) was collected and filtered through 0.2 ?m bottle top filters (Nalgene, Rochester, NY, USA)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	22132229	NA	NA	NA	Gilberto Flores	CCME	GAZ:United States of America	NA	NA	0	NA	NA	NA	NA	n	NA	Noah Fierer	Flores_restroom_surface_biogeography	pyrosequencing	NA	NA	NA	restroom surface metagenome	3	ENVO:city	Titanium	NA	NA	NA	NA	NA	flores_restroom_surface_biogeography	restroom surface biogeography	256318	CCME	NA	NA	NA	1335	"Genomic DNA was extracted from the swabs and filters using the MO BIO PowerSoil DNA isolation kit following the manufacturer's protocol with the modifications of Fierer?et al.?[14]. A portion of the 16 S rRNA gene spanning the V1_V2 regions was amplified using the primer set (27 F/338R), PCR mixture conditions and thermal cycling conditions described in Fierer?et al.?[15]. PCR amplicons of triplicate reactions for each sample were pooled at approximately equal amounts and pyrosequenced at 454 Life Sciences (Branford, CT, USA) on their GS Junior system. A total of 337,333 high-quality partial 16 S rRNA gene sequences were obtained from 101 of the 120 surface samples collected, averaging approximately 3,340 sequences per sample (ranging from 513_6,771) (Table S1) in 4 GS Junior runs, with the best run containing 116,004 high-quality reads. An additional 16,416 sequences (ranging from 2161_5084 per sample) were generated for five of the six water samples collected for source tracking analysis. Each sample was amplified with a unique barcode to enable multiplexing in the GS Junior runs. The barcoded sequencing reads can be separated by data analysis software providing high confidence in assigning sequencing read to each sample. Sequence data generated as part of this study is available upon request by contacting the corresponding author."	NA	NA	y	NA	NA	ENVO:surface	16S rRNA	Porter	NA	V2	CCME	NA	Toilet seat	NA	NA	NA	40.01499	NA	NA	Toilet seat
PTCM5.489525	ATACACGTGGCG	YATGCTGCCTCCCGTAGGAGT	jesse.stombaugh@colorado.edu	NA	restroom surface	Toilet_seat	NA	NA	NA	n	12/13/10	restroom surface metagenome	NA	GFL0IUQ	NA	mimarks-survey	NA	NA	NA	NA	0	NA	NA	1624.1	10-Nov	Male	0	NA	NA	NA	NA	NA	NA	NA	NA	NA	NA	"1,swab"	C	NA	Microbial biogeography of public restroom surfaces	NA	NA	-105.27055	y	NA	CCME	NA	NA	NA	"We spend the majority of our lives indoors where we are constantly exposed to bacteria residing on surfaces. However, the diversity of these surface-associated communities is largely unknown. We explored the biogeographical patterns exhibited by bacteria across ten surfaces within each of twelve public restrooms. Using high-throughput barcoded pyrosequencing of the 16S rRNA gene, we identified 19 bacterial phyla across all surfaces. Most sequences belonged to four phyla: Actinobacteria, Bacteriodetes, Firmicutes and Proteobacteria. The communities clustered into three general categories: those found on surfaces associated with toilets, those on the restroom floor, and those found on surfaces routinely touched with hands. On toilet surfaces, gut-associated taxa were more prevalent, suggesting fecal contamination of these surfaces. Floor surfaces were the most diverse of all communities and contained several taxa commonly found in soils. Skin-associated bacteria, especially the Propionibacteriaceae, dominated surfaces routinely touched with our hands. Certain taxa were more common in female than in male restrooms as vagina-associated Lactobacillaceae were widely distributed in female restrooms, likely from urine contamination. Use of the SourceTracker algorithm confirmed many of our taxonomic observations as human skin was the primary source of bacteria on restroom surfaces. Overall, these results demonstrate that restroom surfaces host relatively diverse microbial communities dominated by human-associated bacteria with clear linkages between communities on or in different body sites and those communities found on restroom surfaces. More generally, this work is relevant to the public health field as we show that human associated microbes are commonly found on restroom surfaces suggesting that bacterial pathogens could readily be transmitted between individuals by the touching of surfaces. Furthermore, we demonstrate that we can use high-throughput analyses of bacterial communities to determine sources of bacteria on indoor surfaces, an approach which could be used to track pathogen transmission and test the efficacy of hygiene practices."	ENVO:building	TCAG	Microbial biogeography of public restroom surfaces	NA	0	"Ten surfaces (door handles into and out of the restroom, handles into and out of a restroom stall, faucet handles, soap dispenser, toilet seat, toilet flush handle, floor around the toilet and floor around the sink) in six male and six female restrooms evenly distributed across two buildings on the University of Colorado at Boulder campus were sampled on a single day in November 2010. Surfaces where sampled using sterile, cotton-tipped swabs as described previously?[14],?[15]. As the 12 restrooms were nearly identical in design, we were able to swab the same area at each location between restrooms. In order to characterize tap water communities as a potential source of bacteria, 1 L of faucet water from six of the restrooms (each building having the same water source for each restroom sampled) was collected and filtered through 0.2 ?m bottle top filters (Nalgene, Rochester, NY, USA)."	FWD:GCCTTGCCAGCCCGCTCAGTCAGAGTTTGATCCTGGCTCAG;REV:TGCTGCCTCCCGTAGGAGT	22132229	NA	NA	NA	Gilberto Flores	CCME	GAZ:United States of America	NA	NA	0	NA	NA	NA	NA	n	NA	Noah Fierer	Flores_restroom_surface_biogeography	pyrosequencing	NA	NA	NA	restroom surface metagenome	3	ENVO:city	Titanium	NA	NA	NA	NA	NA	flores_restroom_surface_biogeography	restroom surface biogeography	256318	CCME	NA	NA	NA	1335	"Genomic DNA was extracted from the swabs and filters using the MO BIO PowerSoil DNA isolation kit following the manufacturer's protocol with the modifications of Fierer?et al.?[14]. A portion of the 16 S rRNA gene spanning the V1_V2 regions was amplified using the primer set (27 F/338R), PCR mixture conditions and thermal cycling conditions described in Fierer?et al.?[15]. PCR amplicons of triplicate reactions for each sample were pooled at approximately equal amounts and pyrosequenced at 454 Life Sciences (Branford, CT, USA) on their GS Junior system. A total of 337,333 high-quality partial 16 S rRNA gene sequences were obtained from 101 of the 120 surface samples collected, averaging approximately 3,340 sequences per sample (ranging from 513_6,771) (Table S1) in 4 GS Junior runs, with the best run containing 116,004 high-quality reads. An additional 16,416 sequences (ranging from 2161_5084 per sample) were generated for five of the six water samples collected for source tracking analysis. Each sample was amplified with a unique barcode to enable multiplexing in the GS Junior runs. The barcoded sequencing reads can be separated by data analysis software providing high confidence in assigning sequencing read to each sample. Sequence data generated as part of this study is available upon request by contacting the corresponding author."	NA	NA	y	NA	NA	ENVO:surface	16S rRNA	Porter	NA	V2	CCME	NA	Toilet seat	NA	NA	NA	40.01499	NA	NA	Toilet seat