Comparison of genetic variation between rare and common congeners of Dipodomys with estimates of contemporary and historical effective population size

Michaela K. Halsey; John D. Stuhler; Natalia J. Bayona-Vásquez; Roy N. Platt II; Jim R. Goetze; Robert E. Martin; Kenneth G. Matocha; Robert D. Bradley; Richard D. Stevens; David A. Ray

doi:10.1371/journal.pone.0274554

Abstract

Species with low effective population sizes are at greater risk of extinction because of reduced genetic diversity. Such species are more vulnerable to chance events that decrease population sizes (e.g. demographic stochasticity). Dipodomys elator, (Texas kangaroo rat) is a kangaroo rat that is classified as threatened in Texas and field surveys from the past 50 years indicate that the distribution of this species has decreased. This suggests geographic range reductions that could have caused population fluctuations, potentially impacting effective population size. Conversely, the more common and widespread D. ordii (Ord’s kangaroo rat) is thought to exhibit relative geographic and demographic stability. We assessed the genetic variation of D. elator and D. ordii samples using 3RAD, a modified restriction site associated sequencing approach. We hypothesized that D. elator would show lower levels of nucleotide diversity, observed heterozygosity, and effective population size when compared to D. ordii. We were also interested in identifying population structure within contemporary samples of D. elator and detecting genetic variation between temporal samples to understand demographic dynamics. We analyzed up to 61,000 single nucleotide polymorphisms. We found that genetic variability and effective population size in contemporary D. elator populations is lower than that of D. ordii. There is slight, if any, population structure within contemporary D. elator samples, and we found low genetic differentiation between spatial or temporal historical samples. This indicates little change in nuclear genetic diversity over 30 years. Results suggest that genetic diversity of D. elator has remained stable despite reduced population size and/or abundance, which may indicate a metapopulation-like system, whose fluctuations might counteract species extinction.

Citation: Halsey MK, Stuhler JD, Bayona-Vásquez NJ, Platt RN II, Goetze JR, Martin RE, et al. (2022) Comparison of genetic variation between rare and common congeners of Dipodomys with estimates of contemporary and historical effective population size. PLoS ONE 17(9): e0274554. https://doi.org/10.1371/journal.pone.0274554

Editor: Natalie Jane Forsdick, Manaaki Whenua: Landcare Research New Zealand, NEW ZEALAND

Received: January 27, 2022; Accepted: August 31, 2022; Published: September 13, 2022

Copyright: © 2022 Halsey et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

Data Availability: Data are available from the NCBI Sequence Read Archive (accession number PRJNA766428).

Funding: DAR, RDS, RDB RFP No. 209f for Endangered Species Research Projects for the Texas Kangaroo Rat Texas State Comptroller https://comptroller.texas.gov/ The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

Competing interests: The authors have declared that no competing interests exist.

Introduction

Measuring genetic variation in rare, threatened, endemic, or endangered species has important implications for management and is integral to conservation efforts [1]. Population genetic summary statistics can be used to delimit management units based on significantly different allele frequencies [2], identify population structure [3], or assess connectivity of demographically disparate subpopulations [4]. One such critical measure for small populations is effective population size, N_e [5], the number of individuals in a population that would resemble an idealized population that would result in the same rate of inbreeding as the study population. This value can be influenced by fluctuations in census size, mating strategy, biased sex ratios, migration, demographic history, spatial dispersion, and population structure [6–9], and typically is far less than the census size. However, the N_e value is often hard to interpret without context. One such context to help understand the potential impacts of fluctuations of N_e is comparison between more restricted, possibly threatened species and a widespread congener, which are presumed to harbor more genetic variation.

The Texas kangaroo rat (Dipodomys elator) is a monotypic, heteromyid rodent that has a limited distribution in north-central Texas [10–14]. Though previously found in two counties in Oklahoma, it appears to have been extirpated from that state [15]. Moreover, D. elator has a small geographic range and a presumably low dispersal capability [16, 17], which increases isolation from nearby subpopulations. D. elator is listed as “vulnerable” by the International Union for Conservation of Nature and Natural Resources [18].

The distribution of the Texas kangaroo rat appears dynamic [19]. For instance, though the species was described from a specimen in Clay County in 1894 [20], it has not been observed there in over 60 years. Resampling of sites where it has been previously documented have failed to detect the species, and new localities of presence have been identified in contemporary surveys [21]. Previous studies of D. elator population genetics have detected population structure and limited diversity [22, 23] and are relevant in assessing genetic diversity within the species. Hamilton et al. 1987 reported low values of genetic distance between D. elator populations in Wichita, Wilbarger, and Hardeman counties and suggested gene flow or short time of separation between populations keeping those values low. Pfau et al. 2019 [23] concluded with the study of mitochondrial DNA and microsatellites, that genetic drift, and not gene flow has had a greater impact on to the genetic structure on the species in Texas, but that the impact of genetic drift was minimal over 17–36 years evaluated. These two studies relied on few molecular markers (i.e. enzymes, mtDNA genes, and microsatellites) but established valuable reference points for our study.

Ord’s kangaroo rat, D. ordii is a medium sized rodent that occurs from Canada into Mexico. Of the 34 subspecies, only one, D. ordii richardsoni, is present in the same region as D. elator [24]. Given its large geographic range and preferred commonly available habitat choice (i.e. sandy soils), D. ordii is not listed on any state or U.S. federal critically threatened and endangered lists. The population in Canada, however, is listed as endangered [25]. To our knowledge, there has not been a range-wide genetic analysis of D. ordii, and the last regional genetic study on D. ordii isoenzymes was published by Beck et al. 1981 [26]. It was found by Beck et al. 1981 [26] that D. ordii in the South Canadian River floodplain in Oklahoma were highly similar in terms of protein variation, with eight of 14 alleles examined being monomorphic.

Sequencing DNA from contemporary samples and from historical museum collections enables comparison of genetic diversity and population structure at different time periods [27, 28]. From such surveys, researchers can gain insight into historical demographics, which can be useful in understanding the underlying factors of genetic diversity. Fortunately, D. elator population surveys are complemented with genetic assays, either from collected specimens [12, 14] or prior genetic analysis [22], setting up critical comparisons of the species across space and time.

Here, we compare population genetic parameters of D. elator with D. ordii and compare D. elator samples from two time periods (1986–1995 and 2015–2017), and for recent samples, investigate differences in genetic diversity across the distribution. We make several predictions: 1) D. elator will exhibit a lower effective population size than D. ordii, and concomitantly, lower nucleotide diversity, lower observed heterozygosity, and higher inbreeding coefficients; 2) there will be greater genetic diversity among contemporary samples of D. elator than in historical samples, as contemporary samples were taken from across the distribution, compared to historical samples collected from three counties in the middle of its distribution; and 3) historical N_e from a coalescent approach for D. ordii and D. elator will demonstrate that D. elator will show a historical decline in N_e, while D. ordii will show stable or increasing N_e.

Methods

Sample collection

Animal handling protocols were approved by the Institutional Animal Care and Use Committee at Texas Tech University (#T14083) and the Texas Parks and Wildlife Department (TPWD #SPR-1013-154). Kangaroo rats were captured using Sherman live traps (23x9x8 cm; H.B. Sherman Traps, Inc. Tallahassee, Florida) during surveys within the historical range of D. elator (Fig 1) from 2015–2017. When a D. elator individual was captured, it was either 1) taken as a voucher specimen for deposition at the Natural Science Research Laboratory (NSRL) at the Museum of Texas Tech University, in which DNA from the liver was used or, 2) had between two to four whiskers extracted from either side of the rostrum [29]. In the latter case, thicker whiskers (i.e., macrovibrissae) were selected with the follicle intact. Whiskers were stored in a sterile vial with 1% sodium dodecyl sulfate (SDS) lysis buffer [30]. A buccal swab was also collected from one D. elator individual as described in detail in Halsey et al. 2021 [29].

Download:

Fig 1. Map of kangaroo rat samples used in this study.

Filled stars indicate contemporary Dipodomys elator samples whereas circles with an ‘x’ are historical D. elator samples. Twenty-eight historical samples in Hardeman are represented by one filled circle with an ‘x’. Filled squares represent D. ordii samples used in the study. Note the contemporary sampling gap located in Foard County, most of Hardman County, and in south Wilbarger County. Trapping restrictions and topography prevented collections in those regions. D. elator is thought to be extirpated in all counties except Cottle, Childress, Hardeman, Wilbarger, and Wichita counties. Map was constructed using ArcMap v10.8.1 and the 1:1,000,000-Scale National Boundaries of the United States data from the USGS EROS (Earth Resources Observatory and Science) Center).

https://doi.org/10.1371/journal.pone.0274554.g001

Other methods of collecting DNA from rats included tail salvages and from toe clips from museum specimens (S1 Table). D. elator tail lengths average about 196 mm [31] and at times the end of the tail (i.e., the plume) was severed by the door of an activated Sherman trap. These salvaged tail plumes were placed in sterile vials of 1% sodium dodecyl sulfate (SDS) lysis buffer [31]. Toe clips samples had been collected in situ from rats from 1986–1995 as part of a population survey of the species by REM and KGM. These samples were not suspended in any buffer and were kept at around -20°C to prevent degradation of DNA.

In total, 70 D. elator samples were analyzed from one of five tissue types (i.e, liver, whisker, tail, buccal swab, and toe clips) and two time periods (historical collections from 1986–1995 and contemporary surveys from 2015–2017; S1 Table and S1 File). D. elator liver samples were collected within the contemporary time frame, and animals were euthanized using isoflurane. Additionally, 26 D. ordii liver samples were collected in five counties from 2015–2017. Contemporary sampling followed guidelines established by the American Society of Mammalogists [32].

Throughout the manuscript the D. elator samples will be referred to using the following descriptors: ‘historical’, from 19861995; ‘contemporary’, from 20152017; ‘west’, collected from Cottle, Childress, or Hardeman counties; and ‘east’ collected from Baylor, Wilbarger, and Wichita counties (Fig 1). Individuals noted as from the “sampling gap” are historical specimens that were captured in an area that is inaccessible or not detected in contemporary surveys.

DNA extraction

DNA was extracted using the Qiagen DNeasy Blood and Tissue spin column protocol (Qiagen; Venlo, Netherlands). For liver, toe clips, and tail salvages, the manufacturer’s recommendations were followed. Using a protocol similar to Moraes-Barros et al. 2007 [33], instruments were autoclaved, and work area was sterilized using 50% bleach and 70% ethanol prior to DNA isolation to prevent contamination. Toes were washed with MilliQ® water. With sterilized scissors, the claw was separated from the rest of the toe and properly discarded. The declawed toe was then placed in a 1.5 mL centrifuge tube with 180 μL buffer TE and 20 μL proteinase-K and incubated overnight. For whisker and buccal swab samples, the protocol found in Halsey et al. 2021 [29] was implemented. In all cases, DNA concentration was fluorometrically quantified using the Qubit 3.0, high sensitivity assay (Invitrogen, Life Technologies, Carlsbad, CA).

3RAD library prep, sequencing, and data processing

RADseq libraries were prepared following the 3RAD protocol found in Bayona-Vásquez et al. 2019 [34]. Details of library prep conditions used in this study are provided in S2 File. In short, restriction enzyme combinations were tested in a subset of samples from both species, and according to digestion patterns and pilot sequence data, the best combination (i.e. MspI, EcoRI, and ClaI), was further used for all samples. DNA were normalized, digested, enzyme-specific adapters were ligated, and ligation products were purified. To generate full- length library constructs, ligated products were amplified using iTru5 and iTru7 primers [35]. A molecular ID tag (iTru 5 8N) was incorporated in the first cycle of PCR for detection of PCR duplicates [34, 36]. PCR products were purified, pooled, and size-selected at a range of 550 bp +/- 15%. Size-selected fragments were purified and sequenced using an Illumina HiSeq 3000 to generate 150 bp paired end data at Oklahoma Medical Research Foundation Genomics Core or Novogene Inc.

Stacks v1.48 and v2.01 [37] was employed to demultiplex, analyze, and export data into other formats. After demultiplexing, poor reads were filtered using the AfterQC ‘after.py’ pipeline [38]. Poor reads were defined as exhibiting a low-quality score (PHRED score < 15), bad overlaps (i.e., mismatched reads), too many ambiguous nucleotides (greater than 40% of the read), short read lengths (< 35 base pairs), or homopolymer regions. If a read failed one of these criteria, it was removed from downstream analyses. Reads (2 x 150 bp) were aligned within Stacks to the D. ordii genome assembly (accession ID GCA_000151885.1; BioProject ID PRJNA20385) using the Burrows-Wheeler aligner [39]. Aligned data are available at The National Center for Biotechnology Information (NCBI) Sequence Read Archive (SRA) under BioProject ID PRJNA766428.

Data were grouped into putative loci, and polymorphisms were identified with the ‘gstacks’ module in Stacks. Common population genetic statistics such as observed and expected heterozygosity, nucleotide diversity, and inbreeding coefficients were calculated using the ‘populations’ module. This step was conducted with four different missingness values (-r) to identify the most appropriate parameter setting. The -r value [37] has been shown to bias population genetic measures, especially in cases where data are not plentiful, and can influence biological implications [40–44]. We used ‘populations’ module using the 75% rule (-r 0.75), two liberal filters (-r of 0.25 and 0.5) and a more conservative filter (-r 0.95). For most downstream analyses, -r 0.75 was used as the main dataset. For other parameters, program defaults were used, including a maximum observed heterozygosity of 1 and a minor allele frequency of 0. These parameters were used for both contemporary and historical samples to reduce bias in summary statistic estimates due to different protocols. Also, as a final filtering step, loci and individuals that had greater than 20% missing data were removed using the ‘missingno’ function in the ‘poppr’ R package [45] for only the -r 0.75 dataset. Code for populations and additional filtering can be found in S3 File.

Population genetics

For genetic diversity, observed and expected heterozygosity were calculated using the ‘summary’ function in the R package adegenet version 2.1.1 [46]. To assess population inbreeding and differentiation, F_IS and F_ST values were calculated using hierfstat [47] and an analysis of molecular variance or AMOVA [48]. Nei’s genetic distances [49] were determined and plotted using the ‘aboot’ function in the poppr R package [45]. Relatedness was calculated using the.relatedness function in vcftools [50].

Estimation of effective population size

To determine effective population size using NeEstimator v2.1 [51], the Genepop file [52] generated by Stacks was used on our contemporary dataset. NeEstimator calculates N_e using three methods: linkage disequilibrium, molecular co-ancestry, and a temporal method. The first two methods were used to determine contemporary effective population sizes per species.

For historical N_e of D. elator, the Extended Bayesian Skyline Plot (EBSP) coalescent test as implemented in BEAST 2.0 [53] was used. Loci containing more than 3 single nucleotide polymorphisms were determined and only individuals with data for these loci were retained. The protocol outlined in Trucchii et al. 2014 [54] was followed, using a strict molecular clock set to 1.0 and a generation time of 3 years [55]. This same process was followed for contemporary D. ordii samples. However, only one individual from Dickens County (Fig 1) was included to avoid misinterpretations due to possible inbreeding since many individuals collected from that county were from a single location.

Population structure

To infer population structure for D. elator, the STRUCTURE algorithm was used [56]. All singletons and private doubletons were removed, which have been shown to mask weak population structure [57, 58]. Only one randomly selected SNP from each locus was used to minimize possible effects of linked data. Allele frequencies were assumed to be correlated, and location prior settings were not used as they did not improve estimates. For all runs, 50,000 burn-in iterations were executed and 200,000 Markov Chain Monte Carlo (MCMC) repetitions with 3 replicates at each K, which ranged from 1–5. We used the Evanno method of delta K to determine the optimal value of “k” [59]. The program DISTRUCT v1.1 was used to visualize the final output of STRUCTURE analyses [60].

Principal components analysis

To visualize genetic structure of the D. elator population without assigning individuals to clusters a priori, a Principal Components Analysis was conducted using the function dudi.pca in the R package ‘adegenet’ version 2.1.1 [46] on historical and contemporary samples. The first two axes explained almost 98% of the data and therefore were the only two axes retained. PCAs for samples separated by time periods are included in S4 File.

Results

In all, 96 kangaroo rats were sequenced and analyzed from two species in eight counties in north-central Texas. Because these data were aligned using the D. ordii genome, it is expected that D. elator will show reduced differentiation estimates and fewer SNPs generated at a lower depth. 3RAD analysis for 70 individual D. elator samples produced over 34 million reads (mean = 5.24x10⁵, SD ± 4.60x10⁵). For D. elator, there were 33 samples for the historical dataset and 37 samples for contemporary dataset. Before filtering within the ‘populations’ module, there were 330,326 SNPs suitable for analysis. Approximately 1.5% of reads per sample were removed from the dataset following AfterQC filtering. After all filtering steps, 3,935 SNPs remained from 55 D. elator individuals (30 contemporary and 25 historical samples).

Similar analysis for the 26 D. ordii samples produced over 10 million reads (mean = 3.62x10⁵, SD ± 4.5x10⁵). Prior to ‘populations’ filtering, approximately 382,514 SNPs were eligible for further analysis. Fewer than 2% of reads per sample were removed from the dataset. Read depth and average number of reads per individual can be found in S2 Table. After all filtering steps, 11,963 SNPs remained from 13 individuals.

Summary population genetics

Across the four -r values evaluated (0.25, 0.5, 0.75, 0.95), there were as few as 7 single nucleotide polymorphisms (SNPs) to as many as 61,000 SNPs to analyze (S3 Table). The general trend was that there were fewer SNPs analyzed as -r value increased and extreme values of -r (i.e. 0.25 and 0.95; S3 Table) yielded stronger deviations between observed and expected heterozygosity across all analyses.

When compared to each other using the -r 0.75 dataset contemporary D. elator samples showed lower levels of observed heterozygosity (0.041, SE±0.0002) than D. ordii (0.158, SE±0.125) which suggests lower genetic diversity in D. elator. Inbreeding coefficient F_IS was positive in all D. elator groups (0.015–0.031) except for in historical samples (-0.007). With an AMOVA, we found that F_ST values demonstrated low but significant genetic differentiation (0.044, p < 0.0001) between pairwise east and west samples. Based on climate, vegetation, edaphic, and land use characteristics across the study area [21], our a priori assumption was that there are two subpopulations (east and west).

Current and historical effective population size

Only the linkage disequilibrium method in NeEstimator v.2.1 produced a value other than ‘Infinite’ for effective population size for D. elator. The estimated N_e of the east group was 171.3, with a 95% confidence interval (CI) of 158–186.9 using the lowest allele frequency (>0). For the west group, both the linkage disequilibrium and molecular co-ancestry methods returned ‘Infinite’ for N_e at all allele frequencies. In the historical Hardeman County dataset, in which all samples were from a single location, we estimate a N_e of 54.8 (CI = 53.7–56.0) using the lowest minor allele frequency (>0). No method with NeEstimator was able to provide an estimate of population size for D. ordii, other than ‘Infinite.’

The Extended Bayesian Skyline plot for D. elator was constructed with 24 individuals and 47 loci. The plot for D. elator showed a decline in effective population size over the last 10,000 years, to an approximate current N_e of 500. For D. ordii, we used 15 individuals and 49 loci. In this case, N_e has increased or remained stable in the last 5,000 years. The current value is estimated to stand at about 20,000 individuals (Fig 2).

Download:

Fig 2.

Extended Bayesian Skyline Plot for Dipodomys elator (top) from 34 individuals and 47 loci and for D. ordii (bottom) from 15 individuals and 49 loci. X-axis is thousands of years ago. Y-axis is effective population size (N_e) in thousands. The y-axis is on a log scale. The dark line on each plot is the mean effective population size, while the shaded gray portions represent the upper highest posterior density (HPD) estimate and the lower highest posterior density (LHD).

https://doi.org/10.1371/journal.pone.0274554.g002

Population substructure

Based on log-likelihood scores (S4 Table) and their respective variances from STRUCTURE, the “best” value for the number of clusters present in the data for D. elator was 3 (Fig 3). For the PCA bi-plot, PC1 accounts for almost 98% of the variation found in the dataset and shows geographic separation along PC2, which only accounts for 0.1% of the variation (Fig 4). Most contemporary samples from the west cluster together and are nested within historical samples, based on Nei’s genetic distance (S1 Fig). The historical PCA for D. elator samples excluding those from the 1960s, which were removed due to poor data quality, shows that all individuals that were taken from the same region (Hardeman County) cluster near the origin and a difference in data quality could result in the three outlying samples (S2 Fig).

Download:

Fig 3. STRUCTURE plot of 60 D. elator samples across three time periods (see text for time breakdown).

The “sampling gap” individuals, those that were found in the areas of Wilbarger and Baylor counties prior to 1980, are completely divergent from later samples;. These results show greater admixture among contemporary subpopulations, though there appears to be some structure in the contemporary east and west samples.

https://doi.org/10.1371/journal.pone.0274554.g003

Download:

Fig 4. Principal components analysis on the genotypes for 55 D. elator samples (historical and contemporary) using the dudi.pca function in R package ‘adegenet’.

While there are no clear clusters emerging on PC1, geographic location seems to correspond with PC2.

https://doi.org/10.1371/journal.pone.0274554.g004

Discussion

This study evaluated changes in genetic diversity across time and space by comparing a rare species with a hypothesized amorphous and restricted distribution to a more common congener with a larger, more defined range. This is only the third population genetic study on Dipodomys elator in over 30 years, and it is the first to make use of genomic techniques, screening from tens to thousands of markers across the genome, making the study valuable for current and future conservation efforts of D. elator. In Hamilton et al. 1987 [22], allozyme markers were used to conclude that there was moderate genetic differentiation among three D. elator localities (Hardeman, Wilbarger, and Wichita counties). This is seemingly incongruent with our results in which we observed little genetic differentiation (F_ST = 0.041), but the difference could simply be the result of the markers used (SNPs versus allozymes). Furthermore, lower heterozygosity in examined nuclear SNPs could reflect the low relative age or variation of the species as most SNPs are likely to be neutral. Similarly, bottleneck event(s) that may have occurred, could have potentially wiped arising genetic variation and homogenized standing populations. Such event was hypothesized in Pfau et al. 2019, but not supported (or detected) by either the estimations of N_e or the Extended Bayesian Skyline Plot method (Fig 2).

More recently, Pfau et al. (2019) [23] observed low mitochondrial DNA variation but high microsatellite diversity within the species and attributed this difference to mtDNA having lower effective population size than neutral nuclear markers such as microsatellites and RAD loci [61]. They concluded that genetic drift and not gene flow has had a greater impact on configuring D. elator genetic diversity, though this impact is minimal. Indeed, genetic drift could play a role in structuring mitochondrial DNA diversity, but more time would be needed to detect reduction of diversity in the nuclear genome using traditional markers such as microsatellites. An insufficient number of polymorphic microsatellite loci limits genetic resolution between individuals with supposed low population-level diversity. Our results suggest that RAD loci can provide greater resolution than with microsatellites when investigating populations with seemingly weak population structure [62].

Together, these three studies, using allozyme, mtDNA, microsatellite, and RAD-Seq markers, offer numerous mean geographic estimates of F_ST within this species. In Hamilton et al. 1987 [22], the mean F_ST was found to be 0.102, Pfau et al. 2019 [23] estimated F_ST of 0.096 from their late 1960s samples, and our study, at the greatest resolution of all previous studies, reveals a mean F_ST value of 0.041. Our lower mean value includes individuals sampled from localities not present in the previous two studies (i.e. Cottle and Childress counties). These results suggest low population differentiation corresponding with geography.

From an overall genetic diversity perspective, our data suggest that there has not been a substantial loss in genetic diversity over the last 30 years, despite what seems to be a decrease in the distribution (and possibly abundance) of Dipodomys elator, similar to what Statham et al. [63] found in D. ingens, the giant kangaroo rat. Like D. elator, the giant kangaroo rat is a species of conservation concern because habitat loss or change has reduced its historical range. So, despite a decline in distribution and census size, which is typically greater than N_e, the genetic diversity of the species is sufficiently high to offset any short-term effects of inbreeding depression. This is supported by our NeEstimator and EBSP estimate of N_e of 170–500, though the NeEstimator value must also be taken under caution because many of the values were ‘Infinite’. This estimate exceeds the recommended minimum value to curtail inbreeding depression, as outlined by the 100/1000 rule [64]. Within this range, there exist enough individuals to mitigate fixation of deleterious alleles due to genetic drift but is low in the longer term (over thousands of years) as suggested by a steady decline in effective population size. In Pfau et al. 2019, it was also found that the N_e of this species was between 65 and 490 individuals.

Results from our contemporary samples confirm that subpopulation differentiation is not substantial (F_ST < 0.05). The STRUCTURE algorithm determined the best value of k to be 3, a finding also described by Pfau et al. 2019, who suggests weak genetic structure overall between east and west samples. The STRUCTURE plot indicates gene flow between hypothesized east and west demes, though it seems unclear exactly how many clusters the contemporary samples may represent. More clusters (i.e. subpopulations), while possible, does not seem parsimonious. Second, newly colonized subpopulations on the fringes of ranges can exhibit lower levels of genetic diversity than expected [65]. For our contemporary samples, this is not the case; the low mean value of F_ST (< 0.05) does not seem to support cluster sizes of k = 4 and k = 5.However, STRUCTURE, the PCA, and Nei’s genetic distances do not support two distinct subpopulations, suggesting there is a fair amount of gene flow in the region.

Our a priori subpopulations display low levels of inbreeding and very little genetic differentiation, suggesting one large interbreeding population, though not necessarily panmictic. Our samples were collected on opposite sides of a cline, separated by a region of inaccessible private land, so it was difficult to determine if the slight differentiation is due to geographic distance or if there is true population substructure and isolation from other habitat patches [66]. We included additional historic samples from specimens collected in the 1960s from areas within this “sampling gap” to answer this question. We anticipated that if the contemporary east and west subpopulations were indeed distinct, then genetic differentiation would be greater between them than to the samples from the sampling gap. In other words, a STRUCTURE plot would show the sampling gap samples as intermediate between the two. Alternatively, if the contemporary east and west subpopulations were considered one population then we would expect greater genetic differentiation between them and our “sampling gap” samples. Our results support the second prediction (Fig 3). However, the periods separating the datasets (anywhere between 20 and 50 years) and the relatively short generation times of kangaroo rats (about 3 years [55]) would lead to high genetic turnover, so these results must be interpreted with caution. If there is substantial genetic turnover, this too could indicate small current effective population size, which supports our estimate of 171–500. Though RAD-Seq and its various types rely on high quality, non-degraded DNA, we were fortunate to obtain enough data to investigate historical genetic diversity of D. elator.

As expected, our D. ordii samples exhibited higher genetic diversity estimates in nearly all categories despite our samples being collected from only five counties in north-central Texas. This emphasizes the substantial genetic diversity and evolutionary potential displayed by the more common D. ordii, compared with its much rarer congener. However, we were surprised to find that D. ordii had a greater inbreeding coefficient than D. elator across some analyses. This pattern can be attributed to sampling bias, given that we sampled from a small portion of the D. ordii range, and half of the D. ordii samples were collected from a single ranch in Dickens County, Texas. However, a relatedness test was performed, and no relatives were identified in the D. ordii set. This is possibly due to the poor data quality of the Dickens County samples (average read depth 1.4; S2 Table). Comparing between individuals from this ranch and a similarly situated subset of D. elator individuals, expected heterozygosity, π, and inbreeding coefficients were largely similar (S5 Table). This suggests that potentially related individuals of D. elator do not show reduced genetic diversity than similarly related D. ordii individuals.

With NeEstimator, we were unable to generate a value of N_e for the current sample of D. ordii, possibly because of the large effective population size of the species [67]. Instead, we used the value calculated from EBSP, which was approximately 20,000 individuals. The plot for D. ordii increased or remained stable, likely as a result of land use changes in the region and perhaps an indication of colonization of new habitat (northward) as glaciers receded after the Last Glacial Maximum 20,000 years ago [68, 69]. The conversion to cultivation, ranching activities, and cattle grazing in this region of Texas have been investigated as factors influencing D. elator populations [19]. The N_e of D. elator declined over time possibly due to land use changes, but this may be an oversimplification.

Coupled with low N_e estimates, and population surveys that recover or fail to locate D. elator in different localities, one possibility is that this population exhibits characteristics of a metapopulation [70, 71]. Metapopulation theory, the study of population dynamics, namely colonization and extinction, of smaller local populations that make up a larger population [71], has been discussed in the context of mammalian conservation biology because it accommodates populations in fragmented habitats [72], but empirical studies to develop metapopulation theory for threatened and endangered mammals are few (see [73, 74]). One reason for the difficulty to meet the original metapopulation criteria [75] is the stringency of the original criteria. In Elmhagen and Angerbjörn [76], the authors relaxed two criteria, adding that subpopulations, not the colonized habitat patch, are the discrete entity, and that these discrete subpopulations differ in their demography, implying asynchronicity. Based on field surveys, analysis of field notes, museum specimens, and species distribution models [21] there is evidence that the D. elator population may benefit from management consideration stemming from metapopulation theory.

However, because this connection to metapopulation theory is still tenuous, the overall population should still be monitored [77] and a long-term demographic study is warranted. Managing the metapopulation must be concerned with maintaining dispersal and gene flow among subpopulations, such as installing or improving habitat corridors or reintroducing D. elator to regions where the species appears extirpated. Should managers elect for more extreme measures to manage D. elator populations, such as captive breeding or population supplementation with the end goal of increasing genetic diversity, knowledge that the population is a metapopulation is critical if managers do not want to inadvertently remove local adaption. Lastly, it is important to note that the metapopulation in a conservation context has several assumptions. One assumption is the “equilibrium” between colonization and extinction across long time scales (i.e. if one patch goes extinct, another is colonized). This seems unlikely in many natural populations [78] but may be possible in D. elator despite its presumed low dispersal capabilities. This type of assumption can be used to appropriately model changes in demography and genetics of D. elator.

There is no lack of research on habitat associations, mainly those evaluating soil and vegetation changes, as they influence D. elator. These studies have greatly improved our understanding of this elusive rodent [16, 17, 19, 79, 80], but we still do not have answers to many basic biological questions. We do know, however, that the population of D. elator seems to track favorable habitat, albeit in a more restricted range than previously recorded [19].

Overall, the population of D. elator exhibits genetic variation lower than that of a species with a predictably greater effective population size. However, contemporary samples show no substantial decrease in genetic diversity from historical samples, suggesting that the D. elator population, though small and constantly shifting, has managed to maintain its genetic diversity.

This study demonstrates the effectiveness of using samples from gradations across the range, rather than at two extremes. Sampling from the extremes of a population range could lead researchers to inappropriate conclusions that could wrongly influence management decisions. Though the current effective population size of D. elator is estimated to be around 171 to 500 individuals, perhaps small population sizes are the status quo for this species. Increasing population size may be unsustainable for this species (greater competition, reduced resources, delayed or forgoing reproduction), but could come at the cost of reduced adaptive potential.

Conservation implications

Researchers interested in natural genetic variation and population structure of mammals should consider the possibility that the population of their organisms of study could be exhibiting the characteristics of a metapopulation. This is especially important for species that are rarely seen or captured. Our findings suggest that the D. elator population could be a metapopulation that must be continuously monitored so that managers can detect immediately any significant losses in genetic diversity and evolutionary potential. Genomic data generated from this study can be used in future investigations. For example, data here can be used to represent another snapshot of the genetic diversity of the species, as the from historical samples served in this study. The data here can be used as a starting point to investigate adaptive loci within the species. Furthermore, given the current advances in molecular techniques and analyses, it is no longer necessary to limit samples in the temporal dimension. Doing so, especially for species that remain understudied, will prove detrimental to any long-term plan for management. We advise, whenever possible, the inclusion of historic and geographically represented samples to fully encapsulate temporal and spatial genetic variability within a possibly imperiled species.

Supporting information

S1 Fig. Nei’s genetic distance dendrogram for 55 D. elator samples (historical and contemporary).

The patchy arrangement of individuals between contemporary spatial demes suggests gene flow between the hypothesized east and west populations, possibly indicative of a metapopulation.

https://doi.org/10.1371/journal.pone.0274554.s001

(TIF)

S2 Fig. Principal components analysis on 25 D. elator historical sample genotypes from Hardeman County using the dudi.pca function in R package ‘adegenet’.

The three samples not clustering with all other is likely a result of different data quality.

https://doi.org/10.1371/journal.pone.0274554.s002

(TIF)

S1 File. Specimens examined section which includes coordinates for each sample used in the study.

Some samples were found on private land, so those coordinates have been withheld.

https://doi.org/10.1371/journal.pone.0274554.s003

(DOCX)

S2 File. 3RAD protocol discussed in more detail.

https://doi.org/10.1371/journal.pone.0274554.s004

(DOCX)

S3 File. Code used to run AfterQC and filtering steps in Stacks and r package poppr.

https://doi.org/10.1371/journal.pone.0274554.s005

(DOCX)

S4 File. Principal components analysis for contemporary D. elator and D. ordii samples, shown separately.

PC1 for the contemporary D. elator samples explained 9.9% of the variation, whereas PC2 accounted for an additional 5.7% of the variation. For historic D. elator samples, PC1 comprised of 6.9% of the variation, and PC2 explained 6.3% of the total variation. PC1 explained about 10% of the variation for contemporary D. ordii samples, and PC2 accounted for 9.8% of the total variation.

https://doi.org/10.1371/journal.pone.0274554.s006

(XLSX)

S1 Table. Seventy Dipodomys elator samples and 26 D. ordii samples used in the genetic analysis including temporal (historical (n = 33), contemporary (n = 37) subpopulations, spatial (east (n = 27) or west (n = 10)) subpopulation, the specific county the individual was found, tissue type, and the museum where the voucher was received.

Museum codes are MSB (Museum of Southwestern Biology), MSU (Midwestern State University), and TTU (Texas Tech University).

https://doi.org/10.1371/journal.pone.0274554.s007

(DOCX)

S2 Table. Statistics on read depth and average number of reads that passed afterQC filtering for 70 D. elator samples and 26 D. ordii individuals.

Sample indicated with asterisk were removed from downstream analyses due to missing data. Samples KR_01, KR_02, KR_04, KR_05, and KR_06 were not part of the initial sequencing but were added later. Raw read number has been included for these 5 samples.

https://doi.org/10.1371/journal.pone.0274554.s008

(DOCX)

S3 Table. Summary statistics calculated in Stacks for 26 D. ordii and 38 D. elator contemporary samples.

Private alleles are those alleles not shared with any other subpopulation.

https://doi.org/10.1371/journal.pone.0274554.s009

(DOCX)

S4 Table. Log-likelihood and delta K values used in the Evanno method for D. elator STRUCTURE analysis.

https://doi.org/10.1371/journal.pone.0274554.s010

(DOCX)

S5 Table. General summary statistics calculated in Stacks for a comparison between 3 individuals from each species that were collected in proximity (i.e. same tract of land).

Private alleles are those alleles not shared with any other subpopulation. Observed and expected heterozygosity are the proportion of loci that are heterozygous based on Hardy-Weinberg frequencies. Historical observed and expected heterozygosity for D. elator are 0.042 and 0.038, respectively. π is a measure of nucleotide diversity. FIS indicates the inbreeding coefficient. The heterozygosity found in the Wichita samples is very large compared to what was found in the species as a whole. This is likely an artefact of which SNPs were used in this comparison. The SNPs evaluated are shared among both D. elator and D. ordii, and D. ordii is a common species, so the SNPs evaluated would be expected to have greater levels of diversity than SNPs evaluated in just D. elator.

https://doi.org/10.1371/journal.pone.0274554.s011

(DOCX)

Acknowledgments

Samples analyzed were loaned from the following collections: Museum of Texas Tech in Lubbock, Texas, Midwestern State University in Wichita Falls, Texas, and the Museum of Southwestern Biology in Albuquerque, New Mexico. Moreover, we would like to acknowledge Travis Glenn for suggesting 3RAD sequencing and guiding the testing of the protocol. This work would not be possible without cooperation from private landowners who allowed us to collect on their land. We sincerely appreciate the efforts of C. Garcia, C. Brothers, A. Kildow, J. Keats, S. de la Piedra, M. Krishnamoorthy, C. Rios-Blanco, G. Langlois, E. Stukenholtz, D. Arenas, L. Lindsay, T. Soniat, E. Roberts, E. Wright, I. Vasquez, J. Francis and many others for their help with fieldwork. Also, we would like to acknowledge the High-Performance Computing Center (HPCC) at Texas Tech University for providing substantial logistical support. We also like to thank H. Wilson and J. Grimshaw and all anonymous reviewers for constructive feedback used to improve earlier versions of this manuscript.

References

1. Frankham R, Ballou SE, Briscoe DA, Ballou JD. Introduction to conservation genetics. Cambridge University Press; 2002 Mar 14.
2. Moritz C. Defining ‘evolutionarily significant units’ for conservation. Trends Ecol Evol. 1994 Oct 1;9(10):373–5. pmid:21236896
- View Article
- PubMed/NCBI
- Google Scholar
3. Rousset F. Genetic structure and selection in subdivided populations (MPB-40). Princeton University Press; 2013 Feb 15.
4. Slatkin M. Gene flow and the geographic structure of natural populations. Science. 1987 May 15;236(4803):787–92. pmid:3576198
- View Article
- PubMed/NCBI
- Google Scholar
5. Wright S. Evolution in Mendelian populations. Genetics. 1931 Mar;16(2):97.
- View Article
- Google Scholar
6. Franklin IR. Evolutionary change in small populations. Conservation biology: an evolutionary-ecological perspective. 1980;395.
- View Article
- Google Scholar
7. Frankham R. Effective population size/adult population size ratios in wildlife: a review. Genet Res. 1995 Oct;66(2):95–107.
- View Article
- Google Scholar
8. Harmon LJ, Braude S. 12. Conservation of Small Populations: Effective Population Sizes, Inbreeding, and the 50/500 Rule. In An introduction to methods and models in ecology, evolution, and conservation biology 2010 Jan 4 (pp. 125–138). Princeton University Press.
9. Wang J, Santiago E, Caballero A. Prediction and estimation of effective population size. Heredity. 2016 Oct;117(4):193–206. pmid:27353047
- View Article
- PubMed/NCBI
- Google Scholar
10. Dalquest WW, Collier G. Notes on Dipodomys elator, a rare kangaroo rat. Southwest Nat. 1964 Aug 20:146–50.
- View Article
- Google Scholar
11. Baccus JT. Mammals of Baylor County, Texas. Texas J Sci. 1971 Jan 1;22(2–3):177.
- View Article
- Google Scholar
12. Martin RE, Matocha KG. Distributional status of the kangaroo rat, Dipodomys elator. J. Mammal. 1972 Nov 1;53(4):873–7.
- View Article
- Google Scholar
13. Carter DC, Jones JK, Jones C, Suttkus RD. Dipodomys elator. Mamm Species. 1985 May 24(232):1–3.
- View Article
- Google Scholar
14. Martin RE, Matocha KG. The Texas kangaroo rat, Dipodomys elator, from Motley Co., Texas, with notes on habitat attributes. Southwest Nat. 1991 Sep 1;36(3):354–6.
- View Article
- Google Scholar
15. Baumgardner GD. A recent specimen of the Texas kangaroo rat, Dipodomys elator (Heteromyidae), from Oklahoma. Southwest Nat. 1987 Jun 15;32(2):285–6.
- View Article
- Google Scholar
16. Garner HW. Population dynamics, reproduction, and activities of the kangaroo rat, Dipodomys ordii, in western Texas. 1970. (Doctoral dissertation, Texas Tech University).
17. Stangl FB, Schafer TS, Goetze JR, Pinchak WOpportunistic use of modified and disturbed habitat by the Texas kangaroo rat (Dipodomys elator). Texas J Sci. 1992 Feb 1;44(1):25–35.
- View Article
- Google Scholar
18. Wahle R, Roth E, Horner P. 2018. Dipodomys elator. The IUCN Red List of Threatened Species 2018: e.T6675A22227507. https://dx.doi.org/10.2305/IUCN.UK.2018-1.RLTS.T6675A22227507.en. Downloaded on 06 September 2021.
- View Article
- Google Scholar
19. Nelson AD, Goetze JR, Watson E, Nelson M. Changes in vegetation patterns and their effect on Texas Kangaroo rats (Dipodomys elator). Texas J Sci. 2009 May 1;61(2):119–30.
- View Article
- Google Scholar
20. Merriam CH. Preliminary descriptions of eleven new kangaroo rats of the genera Dipodomys and Perodipus. Proc Bio Soc Wash. 1894;9:109–16.
- View Article
- Google Scholar
21. Stuhler JD, Halsey MK, Portillo-Quintero C, Ray DA, Bradley RD, Stevens RD. Endangered Species Research: Texas kangaroo rat (Dipodomys elator). Texas Comptroller. 2020.
- View Article
- Google Scholar
22. Hamilton MJ, Chesser RK, Best TL. Genetic variation in the Texas kangaroo rat, Dipodomys elator Merriam. J. Mammal. 1987 Nov 27;68(4):775–81.
- View Article
- Google Scholar
23. Pfau RS, Goetze JR, Martin RE, Matocha KG, Nelson AD. Spatial and temporal genetic diversity of the Texas kangaroo rat, Dipodomys elator (Rodentia: Heteromyidae). J. Mammal. 2019 Jul 27;100(4):1169–81.
- View Article
- Google Scholar
24. Garrison TE, Best TL. Dipodomys ordii. Mamm Species.1990 Apr 26(353):1–0.
- View Article
- Google Scholar
25. Environment Canada. Recovery Strategy for the Ord’s Kangaroo Rat (Dipodomys ordii) in Canada. Species at Risk Act Recovery Strategy Series. Environment Canada, Ottawa. 2012.
26. Beck ML, Kennedy ML, Biggers CJ. Genic variation in Ord’s kangaroo rat Dipodomys ordii in Oklahoma. Am Midl Nat. 1981 Jan 1:77–83.
- View Article
- Google Scholar
27. Wandeler P, Hoeck PE, Keller LF. Back to the future: museum specimens in population genetics. Trends Ecol Evol. 2007 Dec 1;22(12):634–42. pmid:17988758
- View Article
- PubMed/NCBI
- Google Scholar
28. Bradley RD, Bradley LC, Garner HJ, Baker RJ. Assessing the value of natural history collections and addressing issues regarding long-term growth and care. BioSc. 2014 Dec 1;64(12):1150–8.
- View Article
- Google Scholar
29. Halsey MK, Stuhler JD, Platt RN II, Bradley RD, Stevens RD, Ray DA. Cheeky Business: Comparing DNA Yield of Buccal, Fecal, and Whisker Samples for Minimally Invasive Genetic Research. Occassion. Papers, Mus Tex Tech Univ. 2021;374:1–9.
- View Article
- Google Scholar
30. Schmidly DJ, Bradley RD. The mammals of Texas. University of Texas Press; 2016 Aug 9.
31. Longmire JL, Maltbie M, Baker RJ. 1997. Use of" lysis buffer" in DNA isolation and its implication for museum collections. Occasion Papers, Mus Tex Tech Univ. 1997;163: 1–3.
- View Article
- Google Scholar
32. Sikes RS, Animal Care and Use Committee of the American Society of Mammalogists. 2016 Guidelines of the American Society of Mammalogists for the use of wild mammals in research and education. J. Mammal. 2016 Jun 9;97(3):663–88. pmid:29692469
- View Article
- PubMed/NCBI
- Google Scholar
33. Moraes-Barros ND, Morgante JS. A simple protocol for the extraction and sequence analysis of DNA from study skin of museum collections. Genet Mol Bio. 2007 Mar 2;30:1181–5.
- View Article
- Google Scholar
34. Bayona-Vásquez NJ, Glenn TC, Kieran TJ, Pierson TW, Hoffberg SL, Scott PA, et al. Adapterama III: Quadruple-indexed, double/triple-enzyme RADseq libraries (2RAD/3RAD). PeerJ. 2019 Oct 11;7:e7724. pmid:31616583
- View Article
- PubMed/NCBI
- Google Scholar
35. Glenn TC, Nilsen RA, Kieran TJ, Finger JW, Pierson TW, Bentley KE, et al. Adapterama I: universal stubs and primers for thousands of dual-indexed Illumina libraries (iTru & iNext). BioRxiv. 2016 Jan 1:049114.
- View Article
- Google Scholar
36. Hoffberg SL, Kieran TJ, Catchen JM, Devault A, Faircloth BC, Mauricio R, et al. RAD cap: sequence capture of dual‐digest RAD seq libraries with identifiable duplicates and reduced missing data. Mol Ecol Resour. 2016 Sep;16(5):1264–78. pmid:27416967
- View Article
- PubMed/NCBI
- Google Scholar
37. Catchen J, Hohenlohe PA, Bassham S, Amores A, Cresko WA. Stacks: an analysis tool set for population genomics. Mol Ecol. 2013 Jun;22(11):3124–40. pmid:23701397
- View Article
- PubMed/NCBI
- Google Scholar
38. Chen S, Huang T, Zhou Y, Han Y, Xu M, Gu J. AfterQC: automatic filtering, trimming, error removing and quality control for fastq data. BMC Bioinformatics. 2017 Mar;18(3):91–100. pmid:28361673
- View Article
- PubMed/NCBI
- Google Scholar
39. Li H, Durbin R. Fast and accurate long-read alignment with Burrows–Wheeler transform. Bioinform. 2010 Mar 1;26(5):589–95. pmid:20080505
- View Article
- PubMed/NCBI
- Google Scholar
40. Mastretta‐Yanes A, Arrigo N, Alvarez N, Jorgensen TH, Piñero D, Emerson BC. Restriction site‐associated DNA sequencing, genotyping error estimation and de novo assembly optimization for population genetic inference. Mol Ecol Resour. 2015 Jan;15(1):28–41. pmid:24916682
- View Article
- PubMed/NCBI
- Google Scholar
41. Gautier M, Gharbi K, Cezard T, Foucaud J, Kerdelhué C, Pudlo P, et al. The effect of RAD allele dropout on the estimation of genetic variation within and between populations. Mol Ecol. 2013 Jun;22(11):3165–78. pmid:23110526
- View Article
- PubMed/NCBI
- Google Scholar
42. Huang H, Knowles LL. Unforeseen consequences of excluding missing data from next-generation sequences: simulation study of RAD sequences. Syst Biol. 2016 May 1;65(3):357–65. pmid:24996413
- View Article
- PubMed/NCBI
- Google Scholar
43. Hosner PA, Faircloth BC, Glenn TC, Braun EL, Kimball RT. Avoiding missing data biases in phylogenomic inference: an empirical study in the landfowl (Aves: Galliformes). Mol Biol Evol. 2016 Apr 1;33(4):1110–25. pmid:26715628
- View Article
- PubMed/NCBI
- Google Scholar
44. Hodel RG, Chen S, Payton AC, McDaniel SF, Soltis P, Soltis DE. Adding loci improves phylogeographic resolution in red mangroves despite increased missing data: comparing microsatellites and RAD-Seq and investigating loci filtering. Sci Rep. 2017 Dec 14;7(1):1–4.
- View Article
- Google Scholar
45. Kamvar ZN, Tabima JF, Grünwald NJ. Poppr: an R package for genetic analysis of populations with clonal, partially clonal, and/or sexual reproduction. PeerJ. 2014 Mar 4;2:e281. pmid:24688859
- View Article
- PubMed/NCBI
- Google Scholar
46. Jombart T. 2008. adegenet: a R package for the multivariate analysis of genetic markers. Bioinform 24:1403–1405.
- View Article
- Google Scholar
47. Goudet J, Jombart T, Goudet MJ. Package ‘hierfstat’. R package version 0.04‐22. Retrieved from http://www.r-project.org, http://github.com/jgx65/hierfstat. 2015 Nov 24.
48. Excoffier L, Laval G, Schneider S. Arlequin (version 3.0): an integrated software package for population genetics data analysis. Evol Bioinform. 2005 Jan;1:117693430500100003.
- View Article
- Google Scholar
49. Linck E, Battey CJ. Minor allele frequency thresholds strongly affect population structure inference with genomic data sets. Mol Ecol Resour. 2019 May;19(3):639–47. pmid:30659755
- View Article
- PubMed/NCBI
- Google Scholar
50. Danecek P, Auton A, Abecasis G, Albers CA, Banks E, DePristo MA, et al. The variant call format and VCFtools. Bioinform. 2011 Aug;27(15):2156–8. pmid:21653522
- View Article
- PubMed/NCBI
- Google Scholar
51. Do C, Waples RS, Peel D, Macbeth GM, Tillett BJ, Ovenden JR. NeEstimator v2: re‐implementation of software for the estimation of contemporary effective population size (Ne) from genetic data. Mol Ecol Resour. 2014 Jan;14(1):209–14. pmid:23992227
- View Article
- PubMed/NCBI
- Google Scholar
52. Rousset F. genepop’007: a complete re‐implementation of the genepop software for Windows and Linux. Mol Ecol Resour. 2008 Jan;8(1):103–6. pmid:21585727
- View Article
- PubMed/NCBI
- Google Scholar
53. Heled J, Drummond AJ. Bayesian inference of population size history from multiple loci. BMC Evol Biol. 2008 Dec;8(1):1–5. pmid:18947398
- View Article
- PubMed/NCBI
- Google Scholar
54. Trucchi E, Gratton P, Whittington JD, Cristofari R, Le Maho Y, Stenseth NC, et al. King penguin demography since the last glaciation inferred from genome-wide data. Pro Royal Soc B: Biol Science. 2014 Jul 22;281(1787):20140528. pmid:24920481
- View Article
- PubMed/NCBI
- Google Scholar
55. Pacifici M, Santini L, Di Marco M, Baisero D, Francucci L, Marasini GG, et al. Generation length for mammals. Nat Conserv. 2013 Nov 13;5:89.
- View Article
- Google Scholar
56. Pritchard JK, Stephens M, Donnelly P. Inference of population structure using multilocus genotype data. Genetics. 2000 Jun 1;155(2):945–59. pmid:10835412
- View Article
- PubMed/NCBI
- Google Scholar
57. Nei M. Genetic distance between populations. Am Nat. 1972 May 1;106(949):283–92.
- View Article
- Google Scholar
58. Mynhardt S, Bennett NC, Bloomer P. New insights from RADseq data on differentiation in the Hottentot golden mole species complex from South Africa. Mol Phylogenet Evol. 2020 Feb 1;143:106667. pmid:31676418
- View Article
- PubMed/NCBI
- Google Scholar
59. Evanno G, Regnaut S, Goudet J. Detecting the number of clusters of individuals using the software STRUCTURE: a simulation study. Mol Ecol. 2005 Jul;14(8):2611–20. pmid:15969739
- View Article
- PubMed/NCBI
- Google Scholar
60. Rosenberg NA. DISTRUCT: a program for the graphical display of population structure. Mol Ecol Notes. 2004 Mar;4(1):137–8.
- View Article
- Google Scholar
61. Birky CW Jr, Fuerst P, Maruyama T. Organelle gene diversity under migration, mutation, and drift: equilibrium expectations, approach to equilibrium, effects of heteroplasmic cells, and comparison to nuclear genes. Genetics. 1989 Mar 1;121(3):613–27. pmid:2714640
- View Article
- PubMed/NCBI
- Google Scholar
62. Andrews KR, Good JM, Miller MR, Luikart G, Hohenlohe PA. Harnessing the power of RADseq for ecological and evolutionary genomics. Nat Rev Genet. 2016 Feb;17(2):81–92. pmid:26729255
- View Article
- PubMed/NCBI
- Google Scholar
63. Statham MJ, Bean WT, Alexander N, Westphal MF, Sacks BN. Historical population size change and differentiation of relict populations of the endangered giant kangaroo rat. J Hered. 2019 Jul;110(5):548–58. pmid:30715400
- View Article
- PubMed/NCBI
- Google Scholar
64. Frankham R, Bradshaw CJ, Brook BW. Genetics in conservation management: revised recommendations for the 50/500 rules, Red List criteria and population viability analyses. Biol Conserv. 2014 Feb 1;170:56–63.
- View Article
- Google Scholar
65. Eckert CG, Samis KE, Lougheed SC. Genetic variation across species’ geographical ranges: the central–marginal hypothesis and beyond. Mol Ecol. 2008 Mar;17(5):1170–88. pmid:18302683
- View Article
- PubMed/NCBI
- Google Scholar
66. Audzijonyte A, Vrijenhoek RC. When gaps really are gaps: statistical phylogeography of hydrothermal vent invertebrates. Evolution. 2010 Aug;64(8):2369–84. pmid:20298432
- View Article
- PubMed/NCBI
- Google Scholar
67. Waples RS, Do CH. Linkage disequilibrium estimates of contemporary Ne using highly variable genetic markers: a largely untapped resource for applied conservation and evolution. Evol Appl. 2010 May;3(3):244–62. pmid:25567922
- View Article
- PubMed/NCBI
- Google Scholar
68. Jezkova T, Olah‐Hemmings V, Riddle BR. Niche shifting in response to warming climate after the last glacial maximum: inference from genetic data and niche assessments in the chisel‐toothed kangaroo rat (Dipodomys microps). Glob Chang Biol. 2011 Nov;17(11):3486–502.
- View Article
- Google Scholar
69. Jezkova T, Riddle BR, Card DC, Schield DR, Eckstut ME, Castoe TA. Genetic consequences of postglacial range expansion in two codistributed rodents (genus Dipodomys) depend on ecology and genetic locus. Mol Ecol. 2015 Jan;24(1):83–97. pmid:25413968
- View Article
- PubMed/NCBI
- Google Scholar
70. Levins R. Some demographic and genetic consequences of environmental heterogeneity for biological control. Am Entomol. 1969 Sep 1;15(3):237–40.
- View Article
- Google Scholar
71. Hanski I, Gilpin M. Metapopulation dynamics: brief history and conceptual domain. Biol J Linn Soc Lond. 1991 Jan 1;42(1–2):3–16.
- View Article
- Google Scholar
72. Nathan LR, Kanno Y, Vokoun JC. Population demographics influence genetic responses to fragmentation: A demogenetic assessment of the ‘one migrant per generation’rule of thumb. Biol Conserv. 2017 Jun 1;210:261–72.
- View Article
- Google Scholar
73. Gaona P, Ferreras P, Delibes M. Dynamics and viability of a metapopulation of the endangered Iberian lynx (Lynx pardinus). Ecol Monogr. 1998 Aug;68(3):349–70.
- View Article
- Google Scholar
74. Schooley RL, Branch LC. Enhancing the area–isolation paradigm: habitat heterogeneity and metapopulation dynamics of a rare wetland mammal. Ecol Appl. 2009 Oct;19(7):1708–22. pmid:19831065
- View Article
- PubMed/NCBI
- Google Scholar
75. Hanski I. Metapopulation ecology. Oxford University Press; 1999 Mar 18.
76. Elmhagen B, Angerbjörn A. The applicability of metapopulation theory to large mammals. Oikos. 2001 Jul;94(1):89–100.
- View Article
- Google Scholar
77. Lindenmayer DB, Lacy RC. Metapopulation viability of Leadbeater’s possum, Gymnobelideus leadbeateri, in fragmented old‐growth forests. Ecol Appl. 1995 Feb;5(1):164–82.
- View Article
- Google Scholar
78. Akçakaya HR, Mills G, Doncaster CP. The role of metapopulations in conservation. Key topics in conservation biology. 2007:64–84.
- View Article
- Google Scholar
79. Roberts JD, Packard RL. Comments on movements, home range and ecology of the Texas kangaroo rat, Dipodomys elator Merriam. J. Mammal. 1973 Nov 1;54(4):957–62.
- View Article
- Google Scholar
80. Goetze JR, Stasey WC, Nelson AD, Sudman PD. Habitat attributes and population size of Texas kangaroo rats on an intensely grazed pasture in Wichita County, Texas. Texas J Sci. 2007;59(1):11.
- View Article
- Google Scholar

[ref1] 1. Frankham R, Ballou SE, Briscoe DA, Ballou JD. Introduction to conservation genetics. Cambridge University Press; 2002 Mar 14.

[ref2] 2. Moritz C. Defining ‘evolutionarily significant units’ for conservation. Trends Ecol Evol. 1994 Oct 1;9(10):373–5. pmid:21236896
View Article
PubMed/NCBI
Google Scholar

[3] View Article

[4] PubMed/NCBI

[5] Google Scholar

[ref3] 3. Rousset F. Genetic structure and selection in subdivided populations (MPB-40). Princeton University Press; 2013 Feb 15.

[ref4] 4. Slatkin M. Gene flow and the geographic structure of natural populations. Science. 1987 May 15;236(4803):787–92. pmid:3576198
View Article
PubMed/NCBI
Google Scholar

[8] View Article

[9] PubMed/NCBI

[10] Google Scholar

[ref5] 5. Wright S. Evolution in Mendelian populations. Genetics. 1931 Mar;16(2):97.
View Article
Google Scholar

[12] View Article

[13] Google Scholar

[ref6] 6. Franklin IR. Evolutionary change in small populations. Conservation biology: an evolutionary-ecological perspective. 1980;395.
View Article
Google Scholar

[15] View Article

[16] Google Scholar

[ref7] 7. Frankham R. Effective population size/adult population size ratios in wildlife: a review. Genet Res. 1995 Oct;66(2):95–107.
View Article
Google Scholar

[18] View Article

[19] Google Scholar

[ref8] 8. Harmon LJ, Braude S. 12. Conservation of Small Populations: Effective Population Sizes, Inbreeding, and the 50/500 Rule. In An introduction to methods and models in ecology, evolution, and conservation biology 2010 Jan 4 (pp. 125–138). Princeton University Press.

[ref9] 9. Wang J, Santiago E, Caballero A. Prediction and estimation of effective population size. Heredity. 2016 Oct;117(4):193–206. pmid:27353047
View Article
PubMed/NCBI
Google Scholar

[22] View Article

[23] PubMed/NCBI

[24] Google Scholar

[ref10] 10. Dalquest WW, Collier G. Notes on Dipodomys elator, a rare kangaroo rat. Southwest Nat. 1964 Aug 20:146–50.
View Article
Google Scholar

[26] View Article

[27] Google Scholar

[ref11] 11. Baccus JT. Mammals of Baylor County, Texas. Texas J Sci. 1971 Jan 1;22(2–3):177.
View Article
Google Scholar

[29] View Article

[30] Google Scholar

[ref12] 12. Martin RE, Matocha KG. Distributional status of the kangaroo rat, Dipodomys elator. J. Mammal. 1972 Nov 1;53(4):873–7.
View Article
Google Scholar

[32] View Article

[33] Google Scholar

[ref13] 13. Carter DC, Jones JK, Jones C, Suttkus RD. Dipodomys elator. Mamm Species. 1985 May 24(232):1–3.
View Article
Google Scholar

[35] View Article

[36] Google Scholar

[ref14] 14. Martin RE, Matocha KG. The Texas kangaroo rat, Dipodomys elator, from Motley Co., Texas, with notes on habitat attributes. Southwest Nat. 1991 Sep 1;36(3):354–6.
View Article
Google Scholar

[38] View Article

[39] Google Scholar

[ref15] 15. Baumgardner GD. A recent specimen of the Texas kangaroo rat, Dipodomys elator (Heteromyidae), from Oklahoma. Southwest Nat. 1987 Jun 15;32(2):285–6.
View Article
Google Scholar

[41] View Article

[42] Google Scholar

[ref16] 16. Garner HW. Population dynamics, reproduction, and activities of the kangaroo rat, Dipodomys ordii, in western Texas. 1970. (Doctoral dissertation, Texas Tech University).

[ref17] 17. Stangl FB, Schafer TS, Goetze JR, Pinchak WOpportunistic use of modified and disturbed habitat by the Texas kangaroo rat (Dipodomys elator). Texas J Sci. 1992 Feb 1;44(1):25–35.
View Article
Google Scholar

[45] View Article

[46] Google Scholar

[ref18] 18. Wahle R, Roth E, Horner P. 2018. Dipodomys elator. The IUCN Red List of Threatened Species 2018: e.T6675A22227507. https://dx.doi.org/10.2305/IUCN.UK.2018-1.RLTS.T6675A22227507.en. Downloaded on 06 September 2021.
View Article
Google Scholar

[48] View Article

[49] Google Scholar

[ref19] 19. Nelson AD, Goetze JR, Watson E, Nelson M. Changes in vegetation patterns and their effect on Texas Kangaroo rats (Dipodomys elator). Texas J Sci. 2009 May 1;61(2):119–30.
View Article
Google Scholar

[51] View Article

[52] Google Scholar

[ref20] 20. Merriam CH. Preliminary descriptions of eleven new kangaroo rats of the genera Dipodomys and Perodipus. Proc Bio Soc Wash. 1894;9:109–16.
View Article
Google Scholar

[54] View Article

[55] Google Scholar

[ref21] 21. Stuhler JD, Halsey MK, Portillo-Quintero C, Ray DA, Bradley RD, Stevens RD. Endangered Species Research: Texas kangaroo rat (Dipodomys elator). Texas Comptroller. 2020.
View Article
Google Scholar

[57] View Article

[58] Google Scholar

[ref22] 22. Hamilton MJ, Chesser RK, Best TL. Genetic variation in the Texas kangaroo rat, Dipodomys elator Merriam. J. Mammal. 1987 Nov 27;68(4):775–81.
View Article
Google Scholar

[60] View Article

[61] Google Scholar

[ref23] 23. Pfau RS, Goetze JR, Martin RE, Matocha KG, Nelson AD. Spatial and temporal genetic diversity of the Texas kangaroo rat, Dipodomys elator (Rodentia: Heteromyidae). J. Mammal. 2019 Jul 27;100(4):1169–81.
View Article
Google Scholar

[63] View Article

[64] Google Scholar

[ref24] 24. Garrison TE, Best TL. Dipodomys ordii. Mamm Species.1990 Apr 26(353):1–0.
View Article
Google Scholar

[66] View Article

[67] Google Scholar

[ref25] 25. Environment Canada. Recovery Strategy for the Ord’s Kangaroo Rat (Dipodomys ordii) in Canada. Species at Risk Act Recovery Strategy Series. Environment Canada, Ottawa. 2012.

[ref26] 26. Beck ML, Kennedy ML, Biggers CJ. Genic variation in Ord’s kangaroo rat Dipodomys ordii in Oklahoma. Am Midl Nat. 1981 Jan 1:77–83.
View Article
Google Scholar

[70] View Article

[71] Google Scholar

[ref27] 27. Wandeler P, Hoeck PE, Keller LF. Back to the future: museum specimens in population genetics. Trends Ecol Evol. 2007 Dec 1;22(12):634–42. pmid:17988758
View Article
PubMed/NCBI
Google Scholar

[73] View Article

[74] PubMed/NCBI

[75] Google Scholar

[ref28] 28. Bradley RD, Bradley LC, Garner HJ, Baker RJ. Assessing the value of natural history collections and addressing issues regarding long-term growth and care. BioSc. 2014 Dec 1;64(12):1150–8.
View Article
Google Scholar

[77] View Article

[78] Google Scholar

[ref29] 29. Halsey MK, Stuhler JD, Platt RN II, Bradley RD, Stevens RD, Ray DA. Cheeky Business: Comparing DNA Yield of Buccal, Fecal, and Whisker Samples for Minimally Invasive Genetic Research. Occassion. Papers, Mus Tex Tech Univ. 2021;374:1–9.
View Article
Google Scholar

[80] View Article

[81] Google Scholar

[ref30] 30. Schmidly DJ, Bradley RD. The mammals of Texas. University of Texas Press; 2016 Aug 9.

[ref31] 31. Longmire JL, Maltbie M, Baker RJ. 1997. Use of" lysis buffer" in DNA isolation and its implication for museum collections. Occasion Papers, Mus Tex Tech Univ. 1997;163: 1–3.
View Article
Google Scholar

[84] View Article

[85] Google Scholar

[ref32] 32. Sikes RS, Animal Care and Use Committee of the American Society of Mammalogists. 2016 Guidelines of the American Society of Mammalogists for the use of wild mammals in research and education. J. Mammal. 2016 Jun 9;97(3):663–88. pmid:29692469
View Article
PubMed/NCBI
Google Scholar

[87] View Article

[88] PubMed/NCBI

[89] Google Scholar

[ref33] 33. Moraes-Barros ND, Morgante JS. A simple protocol for the extraction and sequence analysis of DNA from study skin of museum collections. Genet Mol Bio. 2007 Mar 2;30:1181–5.
View Article
Google Scholar

[91] View Article

[92] Google Scholar

[ref34] 34. Bayona-Vásquez NJ, Glenn TC, Kieran TJ, Pierson TW, Hoffberg SL, Scott PA, et al. Adapterama III: Quadruple-indexed, double/triple-enzyme RADseq libraries (2RAD/3RAD). PeerJ. 2019 Oct 11;7:e7724. pmid:31616583
View Article
PubMed/NCBI
Google Scholar

[94] View Article

[95] PubMed/NCBI

[96] Google Scholar

[ref35] 35. Glenn TC, Nilsen RA, Kieran TJ, Finger JW, Pierson TW, Bentley KE, et al. Adapterama I: universal stubs and primers for thousands of dual-indexed Illumina libraries (iTru & iNext). BioRxiv. 2016 Jan 1:049114.
View Article
Google Scholar

[98] View Article

[99] Google Scholar

[ref36] 36. Hoffberg SL, Kieran TJ, Catchen JM, Devault A, Faircloth BC, Mauricio R, et al. RAD cap: sequence capture of dual‐digest RAD seq libraries with identifiable duplicates and reduced missing data. Mol Ecol Resour. 2016 Sep;16(5):1264–78. pmid:27416967
View Article
PubMed/NCBI
Google Scholar

[101] View Article

[102] PubMed/NCBI

[103] Google Scholar

[ref37] 37. Catchen J, Hohenlohe PA, Bassham S, Amores A, Cresko WA. Stacks: an analysis tool set for population genomics. Mol Ecol. 2013 Jun;22(11):3124–40. pmid:23701397
View Article
PubMed/NCBI
Google Scholar

[105] View Article

[106] PubMed/NCBI

[107] Google Scholar

[ref38] 38. Chen S, Huang T, Zhou Y, Han Y, Xu M, Gu J. AfterQC: automatic filtering, trimming, error removing and quality control for fastq data. BMC Bioinformatics. 2017 Mar;18(3):91–100. pmid:28361673
View Article
PubMed/NCBI
Google Scholar

[109] View Article

[110] PubMed/NCBI

[111] Google Scholar

[ref39] 39. Li H, Durbin R. Fast and accurate long-read alignment with Burrows–Wheeler transform. Bioinform. 2010 Mar 1;26(5):589–95. pmid:20080505
View Article
PubMed/NCBI
Google Scholar

[113] View Article

[114] PubMed/NCBI

[115] Google Scholar

[ref40] 40. Mastretta‐Yanes A, Arrigo N, Alvarez N, Jorgensen TH, Piñero D, Emerson BC. Restriction site‐associated DNA sequencing, genotyping error estimation and de novo assembly optimization for population genetic inference. Mol Ecol Resour. 2015 Jan;15(1):28–41. pmid:24916682
View Article
PubMed/NCBI
Google Scholar

[117] View Article

[118] PubMed/NCBI

[119] Google Scholar

[ref41] 41. Gautier M, Gharbi K, Cezard T, Foucaud J, Kerdelhué C, Pudlo P, et al. The effect of RAD allele dropout on the estimation of genetic variation within and between populations. Mol Ecol. 2013 Jun;22(11):3165–78. pmid:23110526
View Article
PubMed/NCBI
Google Scholar

[121] View Article

[122] PubMed/NCBI

[123] Google Scholar

[ref42] 42. Huang H, Knowles LL. Unforeseen consequences of excluding missing data from next-generation sequences: simulation study of RAD sequences. Syst Biol. 2016 May 1;65(3):357–65. pmid:24996413
View Article
PubMed/NCBI
Google Scholar

[125] View Article

[126] PubMed/NCBI

[127] Google Scholar

[ref43] 43. Hosner PA, Faircloth BC, Glenn TC, Braun EL, Kimball RT. Avoiding missing data biases in phylogenomic inference: an empirical study in the landfowl (Aves: Galliformes). Mol Biol Evol. 2016 Apr 1;33(4):1110–25. pmid:26715628
View Article
PubMed/NCBI
Google Scholar

[129] View Article

[130] PubMed/NCBI

[131] Google Scholar

[ref44] 44. Hodel RG, Chen S, Payton AC, McDaniel SF, Soltis P, Soltis DE. Adding loci improves phylogeographic resolution in red mangroves despite increased missing data: comparing microsatellites and RAD-Seq and investigating loci filtering. Sci Rep. 2017 Dec 14;7(1):1–4.
View Article
Google Scholar

[133] View Article

[134] Google Scholar

[ref45] 45. Kamvar ZN, Tabima JF, Grünwald NJ. Poppr: an R package for genetic analysis of populations with clonal, partially clonal, and/or sexual reproduction. PeerJ. 2014 Mar 4;2:e281. pmid:24688859
View Article
PubMed/NCBI
Google Scholar

[136] View Article

[137] PubMed/NCBI

[138] Google Scholar

[ref46] 46. Jombart T. 2008. adegenet: a R package for the multivariate analysis of genetic markers. Bioinform 24:1403–1405.
View Article
Google Scholar

[140] View Article

[141] Google Scholar

[ref47] 47. Goudet J, Jombart T, Goudet MJ. Package ‘hierfstat’. R package version 0.04‐22. Retrieved from http://www.r-project.org, http://github.com/jgx65/hierfstat. 2015 Nov 24.

[ref48] 48. Excoffier L, Laval G, Schneider S. Arlequin (version 3.0): an integrated software package for population genetics data analysis. Evol Bioinform. 2005 Jan;1:117693430500100003.
View Article
Google Scholar

[144] View Article

[145] Google Scholar

[ref49] 49. Linck E, Battey CJ. Minor allele frequency thresholds strongly affect population structure inference with genomic data sets. Mol Ecol Resour. 2019 May;19(3):639–47. pmid:30659755
View Article
PubMed/NCBI
Google Scholar

[147] View Article

[148] PubMed/NCBI

[149] Google Scholar

[ref50] 50. Danecek P, Auton A, Abecasis G, Albers CA, Banks E, DePristo MA, et al. The variant call format and VCFtools. Bioinform. 2011 Aug;27(15):2156–8. pmid:21653522
View Article
PubMed/NCBI
Google Scholar

[151] View Article

[152] PubMed/NCBI

[153] Google Scholar

[ref51] 51. Do C, Waples RS, Peel D, Macbeth GM, Tillett BJ, Ovenden JR. NeEstimator v2: re‐implementation of software for the estimation of contemporary effective population size (Ne) from genetic data. Mol Ecol Resour. 2014 Jan;14(1):209–14. pmid:23992227
View Article
PubMed/NCBI
Google Scholar

[155] View Article

[156] PubMed/NCBI

[157] Google Scholar

[ref52] 52. Rousset F. genepop’007: a complete re‐implementation of the genepop software for Windows and Linux. Mol Ecol Resour. 2008 Jan;8(1):103–6. pmid:21585727
View Article
PubMed/NCBI
Google Scholar

[159] View Article

[160] PubMed/NCBI

[161] Google Scholar

[ref53] 53. Heled J, Drummond AJ. Bayesian inference of population size history from multiple loci. BMC Evol Biol. 2008 Dec;8(1):1–5. pmid:18947398
View Article
PubMed/NCBI
Google Scholar

[163] View Article

[164] PubMed/NCBI

[165] Google Scholar

[ref54] 54. Trucchi E, Gratton P, Whittington JD, Cristofari R, Le Maho Y, Stenseth NC, et al. King penguin demography since the last glaciation inferred from genome-wide data. Pro Royal Soc B: Biol Science. 2014 Jul 22;281(1787):20140528. pmid:24920481
View Article
PubMed/NCBI
Google Scholar

[167] View Article

[168] PubMed/NCBI

[169] Google Scholar

[ref55] 55. Pacifici M, Santini L, Di Marco M, Baisero D, Francucci L, Marasini GG, et al. Generation length for mammals. Nat Conserv. 2013 Nov 13;5:89.
View Article
Google Scholar

[171] View Article

[172] Google Scholar

[ref56] 56. Pritchard JK, Stephens M, Donnelly P. Inference of population structure using multilocus genotype data. Genetics. 2000 Jun 1;155(2):945–59. pmid:10835412
View Article
PubMed/NCBI
Google Scholar

[174] View Article

[175] PubMed/NCBI

[176] Google Scholar

[ref57] 57. Nei M. Genetic distance between populations. Am Nat. 1972 May 1;106(949):283–92.
View Article
Google Scholar

[178] View Article

[179] Google Scholar

[ref58] 58. Mynhardt S, Bennett NC, Bloomer P. New insights from RADseq data on differentiation in the Hottentot golden mole species complex from South Africa. Mol Phylogenet Evol. 2020 Feb 1;143:106667. pmid:31676418
View Article
PubMed/NCBI
Google Scholar

[181] View Article

[182] PubMed/NCBI

[183] Google Scholar

[ref59] 59. Evanno G, Regnaut S, Goudet J. Detecting the number of clusters of individuals using the software STRUCTURE: a simulation study. Mol Ecol. 2005 Jul;14(8):2611–20. pmid:15969739
View Article
PubMed/NCBI
Google Scholar

[185] View Article

[186] PubMed/NCBI

[187] Google Scholar

[ref60] 60. Rosenberg NA. DISTRUCT: a program for the graphical display of population structure. Mol Ecol Notes. 2004 Mar;4(1):137–8.
View Article
Google Scholar

[189] View Article

[190] Google Scholar

[ref61] 61. Birky CW Jr, Fuerst P, Maruyama T. Organelle gene diversity under migration, mutation, and drift: equilibrium expectations, approach to equilibrium, effects of heteroplasmic cells, and comparison to nuclear genes. Genetics. 1989 Mar 1;121(3):613–27. pmid:2714640
View Article
PubMed/NCBI
Google Scholar

[192] View Article

[193] PubMed/NCBI

[194] Google Scholar

[ref62] 62. Andrews KR, Good JM, Miller MR, Luikart G, Hohenlohe PA. Harnessing the power of RADseq for ecological and evolutionary genomics. Nat Rev Genet. 2016 Feb;17(2):81–92. pmid:26729255
View Article
PubMed/NCBI
Google Scholar

[196] View Article

[197] PubMed/NCBI

[198] Google Scholar

[ref63] 63. Statham MJ, Bean WT, Alexander N, Westphal MF, Sacks BN. Historical population size change and differentiation of relict populations of the endangered giant kangaroo rat. J Hered. 2019 Jul;110(5):548–58. pmid:30715400
View Article
PubMed/NCBI
Google Scholar

[200] View Article

[201] PubMed/NCBI

[202] Google Scholar

[ref64] 64. Frankham R, Bradshaw CJ, Brook BW. Genetics in conservation management: revised recommendations for the 50/500 rules, Red List criteria and population viability analyses. Biol Conserv. 2014 Feb 1;170:56–63.
View Article
Google Scholar

[204] View Article

[205] Google Scholar

[ref65] 65. Eckert CG, Samis KE, Lougheed SC. Genetic variation across species’ geographical ranges: the central–marginal hypothesis and beyond. Mol Ecol. 2008 Mar;17(5):1170–88. pmid:18302683
View Article
PubMed/NCBI
Google Scholar

[207] View Article

[208] PubMed/NCBI

[209] Google Scholar

[ref66] 66. Audzijonyte A, Vrijenhoek RC. When gaps really are gaps: statistical phylogeography of hydrothermal vent invertebrates. Evolution. 2010 Aug;64(8):2369–84. pmid:20298432
View Article
PubMed/NCBI
Google Scholar

[211] View Article

[212] PubMed/NCBI

[213] Google Scholar

[ref67] 67. Waples RS, Do CH. Linkage disequilibrium estimates of contemporary Ne using highly variable genetic markers: a largely untapped resource for applied conservation and evolution. Evol Appl. 2010 May;3(3):244–62. pmid:25567922
View Article
PubMed/NCBI
Google Scholar

[215] View Article

[216] PubMed/NCBI

[217] Google Scholar

[ref68] 68. Jezkova T, Olah‐Hemmings V, Riddle BR. Niche shifting in response to warming climate after the last glacial maximum: inference from genetic data and niche assessments in the chisel‐toothed kangaroo rat (Dipodomys microps). Glob Chang Biol. 2011 Nov;17(11):3486–502.
View Article
Google Scholar

[219] View Article

[220] Google Scholar

[ref69] 69. Jezkova T, Riddle BR, Card DC, Schield DR, Eckstut ME, Castoe TA. Genetic consequences of postglacial range expansion in two codistributed rodents (genus Dipodomys) depend on ecology and genetic locus. Mol Ecol. 2015 Jan;24(1):83–97. pmid:25413968
View Article
PubMed/NCBI
Google Scholar

[222] View Article

[223] PubMed/NCBI

[224] Google Scholar

[ref70] 70. Levins R. Some demographic and genetic consequences of environmental heterogeneity for biological control. Am Entomol. 1969 Sep 1;15(3):237–40.
View Article
Google Scholar

[226] View Article

[227] Google Scholar

[ref71] 71. Hanski I, Gilpin M. Metapopulation dynamics: brief history and conceptual domain. Biol J Linn Soc Lond. 1991 Jan 1;42(1–2):3–16.
View Article
Google Scholar

[229] View Article

[230] Google Scholar

[ref72] 72. Nathan LR, Kanno Y, Vokoun JC. Population demographics influence genetic responses to fragmentation: A demogenetic assessment of the ‘one migrant per generation’rule of thumb. Biol Conserv. 2017 Jun 1;210:261–72.
View Article
Google Scholar

[232] View Article

[233] Google Scholar

[ref73] 73. Gaona P, Ferreras P, Delibes M. Dynamics and viability of a metapopulation of the endangered Iberian lynx (Lynx pardinus). Ecol Monogr. 1998 Aug;68(3):349–70.
View Article
Google Scholar

[235] View Article

[236] Google Scholar

[ref74] 74. Schooley RL, Branch LC. Enhancing the area–isolation paradigm: habitat heterogeneity and metapopulation dynamics of a rare wetland mammal. Ecol Appl. 2009 Oct;19(7):1708–22. pmid:19831065
View Article
PubMed/NCBI
Google Scholar

[238] View Article

[239] PubMed/NCBI

[240] Google Scholar

[ref75] 75. Hanski I. Metapopulation ecology. Oxford University Press; 1999 Mar 18.

[ref76] 76. Elmhagen B, Angerbjörn A. The applicability of metapopulation theory to large mammals. Oikos. 2001 Jul;94(1):89–100.
View Article
Google Scholar

[243] View Article

[244] Google Scholar

[ref77] 77. Lindenmayer DB, Lacy RC. Metapopulation viability of Leadbeater’s possum, Gymnobelideus leadbeateri, in fragmented old‐growth forests. Ecol Appl. 1995 Feb;5(1):164–82.
View Article
Google Scholar

[246] View Article

[247] Google Scholar

[ref78] 78. Akçakaya HR, Mills G, Doncaster CP. The role of metapopulations in conservation. Key topics in conservation biology. 2007:64–84.
View Article
Google Scholar

[249] View Article

[250] Google Scholar

[ref79] 79. Roberts JD, Packard RL. Comments on movements, home range and ecology of the Texas kangaroo rat, Dipodomys elator Merriam. J. Mammal. 1973 Nov 1;54(4):957–62.
View Article
Google Scholar

[252] View Article

[253] Google Scholar

[ref80] 80. Goetze JR, Stasey WC, Nelson AD, Sudman PD. Habitat attributes and population size of Texas kangaroo rats on an intensely grazed pasture in Wichita County, Texas. Texas J Sci. 2007;59(1):11.
View Article
Google Scholar

[255] View Article

[256] Google Scholar

Figures

Abstract

Introduction

Methods

Sample collection

DNA extraction

3RAD library prep, sequencing, and data processing

Population genetics

Estimation of effective population size

Population structure

Principal components analysis

Results

Summary population genetics

Current and historical effective population size

Population substructure

Discussion

Conservation implications

Supporting information

S1 Fig. Nei’s genetic distance dendrogram for 55 D. elator samples (historical and contemporary).

S2 Fig. Principal components analysis on 25 D. elator historical sample genotypes from Hardeman County using the dudi.pca function in R package ‘adegenet’.

S1 File. Specimens examined section which includes coordinates for each sample used in the study.

S2 File. 3RAD protocol discussed in more detail.

S3 File. Code used to run AfterQC and filtering steps in Stacks and r package poppr.

S4 File. Principal components analysis for contemporary D. elator and D. ordii samples, shown separately.

S2 Table. Statistics on read depth and average number of reads that passed afterQC filtering for 70 D. elator samples and 26 D. ordii individuals.

S3 Table. Summary statistics calculated in Stacks for 26 D. ordii and 38 D. elator contemporary samples.

S4 Table. Log-likelihood and delta K values used in the Evanno method for D. elator STRUCTURE analysis.

S5 Table. General summary statistics calculated in Stacks for a comparison between 3 individuals from each species that were collected in proximity (i.e. same tract of land).

Acknowledgments

References