http://orcid.org/0000-0002-6318-7234
Figures
Abstract
Pod number is an important factor that influences yield in soybean. Here, we used two associated recombinant inbred line (RIL) soybean populations, RIL3613 (containing 134 lines derived from Dongnong L13 × Heihe 36) and RIL6013 (composed of 156 individuals from Dongnong L13 × Henong 60), to identify quantitative trait loci (QTLs) regulating the vertical distribution and quantity of seeds and seed pods. The numbers of pods were quantified in the upper, middle, and lower sections of the plant, as well as in the plants as a whole, and QTLs regulating these spatial traits were mapped using an inclusive complete interval mapping method. A total of 21 and 26 QTLs controlling pod-number-related traits were detected in RIL3613 and RIL6013, respectively, which explained 1.25–11.6698% and 0.0001–7.91% of the phenotypic variation. A total of 34 QTLs were verified by comparison with previous research, were identified in both populations, or were found to regulate multiple traits, indicating their authenticity. These results enhance our understanding of the vertical distribution of pod-number-related traits and support molecular breeding for seed yield.
Citation: Ning H, Yuan J, Dong Q, Li W, Xue H, Wang Y, et al. (2018) Identification of QTLs related to the vertical distribution and seed-set of pod number in soybean [Glycine max (L.) Merri]. PLoS ONE 13(4): e0195830. https://doi.org/10.1371/journal.pone.0195830
Editor: David A. Lightfoot, College of Agricultural Sciences, UNITED STATES
Received: December 7, 2017; Accepted: April 1, 2018; Published: April 17, 2018
Copyright: © 2018 Ning 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: All relevant data are within the paper and its Supporting Information files.
Funding: Natural Science Foundation of Heilongjiang Province [China] (LC201610) to H.N. and (C2015007) to W-X.L., and the Scientific and Technological Foundation of Education Department of Heilongjiang Province (12541z001) to H.N, grants from the National Key Research and Development Program of China (2017YFD0101300) to H.N.
Competing interests: The authors have declared that no competing interests exist.
Abbreviations: LG, Linkage group; PNUA, Number of pods containing one seed in the upper part of the plant; PNMA, Number of pods containing one seed in the middle plant section; PNBA, Number of pods containing one seed in the lower part of the plant; PNUB, Number of pods containing two seeds in the upper part of the plant; PNMB, Number of pods containing two seeds in the middle part of the plant section; PNBB, Number of pods containing two seeds in the lower part of the plant; PNUC, Number of pods containing three seeds in the upper part of the plant; PNMC, Number of pods containing three seeds in the middle plant section; PNBC, Number of pods containing three seeds in the lower part of the plant; PNUD, Number of pods containing four seeds in the upper part of the plant; PNMD, Number of pods containing four seeds in the middle plant section; PNBD, Number of pods containing four seeds in the lower part of the plant; TPA, Total number of pods containing one seed; TPB, Total number of pods containing two seeds; TPC, Total number of pods containing three seeds; TPD, Total number of pods containing four seeds; QTL, Quantitative trait locus; ICIM, Inclusive complete interval mapping method
Introduction
The number of pods per plant is one of the most agronomically important traits in soybean [Glycine max (L.) Merri], and is strongly positively correlated with yield [1]. Numerous studies have mapped quantitative trait loci (QTLs) for pod number, with the aim of increasing the efficiency of breeding for higher yields [2–18]. Previously, 15 QTLs for total pod number per plant (TPNPP) were identified using recombinant inbred lines (RILs) derived from a cross between BARC-8 and Garimpo [2], while another study identified 12 TPNPP QTLs on chromosomes B1, C2, D1a, F, J, and N using a F2:10 RIL population derived from a cross between Charleston and Dongnong 594 [3]. A further nine pod-number QTLs, including two QTLs for the number of pods containing one seed (TPA), one QTL for the number of pods containing two seeds (TPB), two QTLs for three-seed pods (TPC), and four QTLs for the number of four-seed pods (TPD), were identified from a population of introgression lines derived from the donors Harosoy and Clark and the receptor Hongfeng 11 [5]. A similar study using a RIL population of 165 individuals found 11 pod-number QTLs, including one QTL for TPA, five QTLs for TPB, two QTLs for TPC, and three QTLs for TPD [6]. In all, nine QTLs for TPA, six QTLs for TPB, three QTLs for TPC, fifteen QTLs for TPD, and 55 QTLs for TPNPP have been mapped on chromosomes 2–11, 13, and 15–20 (S1 Table).
Soybean pod numbers are also affected by their vertical spatial distribution. The pod numbers in the center of the main stem account for the majority of the pods on each plant, and the seed number in each pod is typically greater in this central region than in pods formed lower on the main stem [19]. In addition, more pods are formed on the upper and central sections of the stem than on the lower sections [20]. These previous findings reveal that pod and seed numbers are not uniform throughout the plant, and understanding the distribution of pods and seeds would be beneficial to improving soybean production.
Previous QTL studies analyzing various aspects of pod numbers in soybean ignored the uneven distribution of pods across the upper, middle, and lower parts of the plant. Physiological research in soybean has revealed that programs to increase pod numbers should consider the differences between the different regions of the plant; therefore, it is imperative to understand the genetic basis of the vertical distribution of pods. In the present study, two soybean RIL populations with a common female parent were used to detect QTLs regulating pod numbers in the upper, middle, and lower parts of the plant. The objective of this paper is to explore the genetic regulation of vertical pod distribution and to identify the QTLs with the largest effects for use in molecular breeding.
Materials and methods
Plant materials
Three soybean varieties, Dongnong L13, Henong 60, and Heihe 36, were employed to construct two RIL populations. These three parents derived from germplasms with extensive genetic differences and highly variable pod number traits (Table 1). Two crosses between Dongnong L13 × Henong 60 and between Dongnong L13 × Heihe 36 were conducted in 2008 (E126.63°, N45.75°) in Harbin, Heilngjiang, China, and 15 and 22 hybrid seeds were harvested for the two crosses, respectively. The F1 seeds were sown in Yacheng (E109.00°, N17.5°) in Hainan Province, China, and the mature plants were harvested in the winter. The non-hybrid seeds were removed following a comparison with the female parent, and the remaining seeds were continuously self-crossed for five generations from 2010 to 2013, in Harbin in the summer and Yacheng in the winter, with each individual selected from single-seed descent. A total of 134 and 156 RILs were ultimately obtained for the two populations, named RIL3613 (Dongnong L13 × Heihe 36) and RIL6013 (Dongnong L13 × Henong 60) respectively, which were used for the construction of the genetic linkage map and QTL mapping in the present study.
Field experiment
The parental lines and RILs were planted in Harbin in 2015, and were grown in a randomized complete block design with three replications. Each plot contained three 3-m rows that were 70 cm apart, and the seeds of an individual line were sown at 6-cm intervals. The field experiment was managed identically to the local soybean crops.
Trait evaluations
Five mature plants were selected randomly from the middle row of each plot and the number of pods containing one, two, three, and four seeds on each node of the main stem were recorded. The number of nodes on the main stem was divided by three, with the top, central, and lower thirds of the plants labeled as the upper, middle, and lower sections of the plant, respectively. If dividing the number of nodes by three left a remainder of 1, the extra node was allocated to the middle section. If the remainder was two, one additional node was allocated to the top section and the other was allocated to the lower section.
For the upper section, the total number of pods containing one, two, three, and four seeds were recorded as PNUA, PNUB, PNUC, and PNUD, respectively. In the middle section, these categories were labeled PNMA, PNMB, PNMC, and PNMD, and in the lower part of the plant they were recorded as PNBA, PNBB, PNBC, and PNBD. The total number of pods containing one, two, three, or four seeds on all nodes of the plant were recorded as TPA, TPB, TPC, and TPD, respectively.
SSR marker analysis
Juvenile leaves were collected from the two RIL populations, frozen in liquid nitrogen, then immediately ground into powder. Total genomic DNA was extracted using the CTAB method [21], and eluted in 50 μl deionized water. Its concentration was determined using a UV752N spectrophotometer (Shanghai Jingke Science Instrument Co. Ltd.) and was diluted to 100 ng–1 in deionized water.
A total of 560 simple sequence repeat (SSR) markers evenly distributed across the soybean genome [22] were selected to screen for polymorphisms between the two parents of each RIL population. Of these, 137 and 150 primer pairs showed polymorphisms in RIL6013 and RIL3613, respectively, and were therefore used for SSR genotyping. An optimized PCR was performed using a total reaction volume of 20 μl, including 3 μl genomic DNA, 2 μl reaction buffer, 3 μl SSR primer, 0.3 μl dNTP, 0.2 μl Taq DNA polymerase, and 11.5 μl ddH2O. The reaction was performed at 94°C for 10 min; followed by 38 cycles of 94°C for 30 s, 50°C for 30 s, and 72°C for 30 s; with a final extension at 72°C for 10 min. A 6% denaturing polyacrylamide gel electrophoresis was used for silver staining, water extraction, development, and genotyping.
Construction of the linkage map and QTL analysis
The linkage maps were constructed using the QTLIciMapping 4.0 software (www.isbreeding.net), using the default setting for all parameters.
The average number of pods from five individuals per line was used for statistical analysis. The inclusive compositive interval mapping method (ICIM) [23], compositive interval mapping method (CIM) and single marker analysis (SMA) [24]. The significant threshold of LOD (logarithm of odds) score for ICIM, LR (likelihood ratio) for CIM and probability over F value for SMA were set as 2.5, 11.5 and 0.05, respectively. When the QTL were detected simultaneously by over two methods, and PVE (phenotypic variation explanation ratio) for ICIM over 2% or R2 (coefficient of determination) for CIM over 0.1, we declared the presence of the QTL. ICIM was implemented by the IciMapping 4.0 software (www.isbreeding.net), and CIM and SMA were implemented by the WinQTLCart 2.5 software (https://brcwebportal.cos.ncsu.edu/qtlcart/WQTLCart.htm).
All data obtained from experiment were listed in S1 Dataset.
Results
Phenotypic analysis
For all 16 pod number traits, there was a large variation among the RIL3613 and RIL6013 lines (Table 1 and S1 Fig); therefore, the two RIL populations were suitable for use to detect QTLs for PNUA, PNMA, PNBA, PNUB, PNMB, PNBB, PNUC, PNMC, PNBC, PNUD, PNMD, PNBD, TPA, TPB, TPC, and TPD. In terms of the pod number-related seed set traits, those related to three seeds per pod (PNUC, PNMC, PNBC) showed the largest means and ranges in both populations, while those related to pods with one seed (PNUA, PNMA, PNBA) exhibited the lowest mean and ranges in both populations. PNUD, PNMD, PNBD showed the second largest range in RIL3613, while PNUB, PNMB, PNBB had second biggest range in RIL6013. Importantly, there were parallel differences between populations in vertical distribution of numbers of pods, with the mean and range of middle part being the largest, and those of the upper and lower regions being approximately equal. In all, these data showed that there was high variation by pod number in seed set and vertical distribution, and different QTLs associated with pod number could likely be identified based on seed set and vertical distribution.
Linkage map
A total of 150 and 137 SSR markers were anchored across all 20 soybean chromosomes in RIL3613 and RIL6013, with total linkage map lengths of 2849.54 cM and 1886.8 cM, respectively. The mean interval lengths for RIL3613 and RIL6013 populations were 21.92 cM and 16.13 cM, respectively. For RIL3613, the length of each linkage group ranged from 1.15 cM to 283.42 cM, with 31 intervals (23.85%) shorter than 10 cM, and 20 intervals (15.38%) longer than 25 cM. For RIL6013, the linkage group length varied from 19.68 cM to 163.67 cM, with 30 intervals (26.50%) shorter than 10 cM, and 14 intervals (11.97%) longer than 25 cM (S2 Table, S2 and S3 Figs).
QTL mapping of the pod-number traits
A total of 47 QTLs were found to be associated with PNUA, PNUD, PNMA, PNMB, PNMC, PNMD, PNBA, PNBB, PNBC, PNBD, TPA, TPB, TPC, and TPD (Table 2). Of these, 21 and26 were identified from RIL3613 and RIL6013, respectively.
QTLs for the numbers of one-seed pods.
Two QTLs for PNUA (qPN-C1-3, and qPN-I-1) were detected in RIL3613. At these two QTLs, the alleles from Heihe 36 could increase PNUA. In RIL6013, seven PNUA QTLs (qPN-D1a-1, qPN-D1b-1, qPN-N-1, qPN-O-2, qPN-B1-1, qPN-J-2, and qPN-G-3) were detected, The additive effects of these seven QTLs were negative, indicating that the alleles that increased PNUA derived from Henong 60.
A total of five QTLs for PNMA (qPN-A1-1, qPN-C2-1, qPN-F-1, qPN-D2-2, and qPN-L-1), located in linkage groups (LGs) A1, C2, A2, F, D2 and L, were detected in RIL3613. The alleles that increased PNMA were carried by Heihe 36. In RIL6013, three QTLs (qPN-D1b-2, qPN-E-1 and qPN-I-2) underlying PNMA were detected in LGs D1b, E and I,. Alleles expressing positive additive effects on PNMA were derived from Henong 60 at all three of these QTLs.
Two QTLs (qPN-M-1 and qPN-I-3) controlling 4.65% and 2.16% of the phenotypic variation in PNBA were detected in RIL3613 in LGs M, and I. The Dongnong L13 allele for qPN-M-1 and the Heihe 36 alleles for qPN-I-3 enhanced PNBA. Six QTLs associated with PNBA (qPN-N-1, qPN-C1-2, qPN-B1-1, qPN-H-1, and qPN-E-3) were detected in LGs N, C1, B1, H, and E in RIL6013. The Henong 60 alleles for five QTLs (qPN-N-1, qPN-C1-2, qPN-B1-1, and qPN-E-3) and the qPN-H-1 allele from Dongnong L13 increased PNBA.
Four QTLs (qPN-D1b-3, qPN-C1-1, qNP-M-1 and qNP-I-3) for TPA were located in LGs D1b, C1, M and I in RIL3613. The additive effects of QTL qPN-D1b-3, qPN-C1-1, and qNP-I-3 were negative, indicating that the alleles carried by Heihe 36 could improve TPA. The additive effects of QTL qPN-M-1 were positive which showed that the alleles carried by Dongnong L13 could improve TPA.
QTLs for the numbers of two-seed pods.
Two QTLs (qPN-M-2 and qPN-O-4) for PNMB were detected in RIL3613, which explained 2.12% and 10.70% of the phenotypic variation, respectively. Synergistic alleles for qPN-O-4 and qPN-M-2 for PNMB were carried by Dongnong L13 and Heihe 36, respectively. One QTLs (qPN-G-2) for PNMB were detected in RIL6013, in which synergistic allele were carried by Henong 60. Two QTLs (qPN-C1-1 and qPN-O-3) associated with PNBB were discovered in RIL3613, which explained 3.30% and 11.67% of the phenotypic variation, respectively. The qPN-O-3 allele with a positive additive effect on PNBB was carried by Dongnong L13, while that of qPN-C1-1 derived from Heihe 36.
One TPB QTLs (qPN-G-2) were detected in RIL6013, and explained 12% (R2) of the phenotypic variation. The Henong 60 allele for qPN-G-2 and the Dongnong L13 allele for qPN-O-4 increased TPB.
QTLs for the numbers of three-seed pods.
Two QTL (qPN-D1b-2 and qPN-L-4) explained 7.91% and 11% of the PNMC phenotypic variation in RIL6013, and the allele providing the positive additive effect was derived from Henong 60 and Dongnong L13, respectively. Three PNBC QTLs (qPN-B1-1, qPN-J-1, and qPN-L-3) detected in RIL6013, and the alleles of these three QTLs that could increase PNBC were derived from Henong 60.
One TPC QTL (qPN-G-2) was detected in RIL6013, and explained just 13% (R2) of the phenotypic variation. The negative effect of this QTL demonstrated that the Henong 60 allele for this QTL increased TPC.
QTLs for the numbers of four-seed pods.
In RIL6013, six QTLs associated with PNUD (qPN-B1-3, qPN-B1-4, qPN-E-3, qPN-G-3, qPN-L-3, and qPN-I-2) were identified, and the Henong 60 alleles for all 14 of these QTLs increased PNUD.
Five QTLs (qPN-D1a-2, qPN-A1-2, qPN-O-1, qPN-B2-1 and qPN-E-2) for PNMD detected in RIL3613. The Heihe 36 alleles for qPN-D1a-2, qPN-O-1, and qPN-B2-1, the Dongnong L13 allele of qPN-A1-2 and qPN-E-2, increased PNMD. In RIL6013, nine QTLs associated with PNMD (qPN-N-1, qPN-O-2, qPN-B1-2, qPN-B1-3, qPN-B1-1, qPN-B1-4, qPN-G-4, qPN-L-3, and qPN-I-2), and the additive effects of these nine QTLs were negative, indicating that the alleles that increase PNMD derived from Henong 60.
Four QTLs underlying PNBD (qPN-A1-2, qPN-B2-1, qPN-D2-1, and qPN-D2-2) were detected in RIL3613. Alleles of qPN-A1-2 and qPN-D2-1 from Dongnong L13 and those of qPN-B2-1, and qPN-D2-2 from Heihe 36 improved PNBD. A total of 13 PNBD QTLs (qPN-D1a-1, qPN-D1b-2, qPN-N-1, qPN-C2-2, qPN-C2-3, qPN-A2-1, qPN-K-1, qPN-O-2, qPN-B1-2, qPN-B1-4 qPN-G-4 and qPN-L-2) were found in RIL6013, and the Henong 60 alleles for all 13 of these QTLs increased PNBD.
Three TPD QTLs (qPN-D1a-2, qPN-B2-1 and qPN-G-1) were detected in LGs D1a, B2 andG in RIL3613. Alleles of these three QTLs from Heihe 36, increased TPD. Eight QTLs (qPN-N-1, qPN-O-2, qPN-B1-2, qPN-B1-3, qPN-B1-4, qPN-G-4, qPN-L-3, and qPN-I-2) underlying TPD were detected in RIL6013 in LGs B1, G, L, O, N, and I, and the negative additive effects of these eight QTLs indicated that the alleles from Henong 60 increased TPD.
Discussion
Analyzing pod numbers in three sections of the plant increased QTL detection power and could facilitate yield improvement
Most genetic and breeding research has focused on the total pod number of entire plants [2–18], ignoring the impact of the vertical distribution of pods. In this research, the pod number of the entire plant was divided into three vertical sections, which enhanced the power of QTL detection. Only 16 QTLs for TPA, TPB, TPC, and TPD were detected in the two populations; however, when the QTL analysis was conducted on the three distinct sections of the plants, 31 other QTLs for pod number were identified. Furthermore, this result elucidated the molecular basis underlying the phenotypic variation in different types of pod in the different sections of the plant, which could enable breeders to improve seed set, by combining desirable genotypes controlling seeds per pod and pod number in the various parts of the plant. Our findings highlight the importance of distinguishing the genetic regulation of seed set and pod numbers in different regions of the plant. Thus, it is necessary to map QTLs by examining the pod numbers separately in upper, middle and lower parts of the plant.
Identification of QTL alleles for use in molecular breeding
Mapping the QTLs that regulate the spatial distribution of pods will enable the practical improvement of seed yields, as breeders can combine QTLs controlling pod numbers in different areas of the plant. The favorable allelic genotypes should be transposed from specific parents to the offspring. In this research, we identified 21 and 26 QTLs associated with pod-number-related traits in RIL3613 and RIL6013, respectively. For RIL3613, QTLs with the favorable genotypes derived from Dongnong L13 include qPN-A1-2, qPN-M-1, qPN-O-3, qPN-O-3, PN-E-2 and qPN-D2-1; and those from Heihe 36 include qPN-D1a-2, qPN-D1b-3, qPN-C1-1, qPN-C1-3, qPN-A1-1, qPN-C2-1, qPN-M-2, qPN-O-1, qPN-F-1, qPN-B2-1, qPN-D2-2, qPN-G-1, qPN-L-1, qPN-I-1 and qPN-I-3. In the RIL6013 population, the QQ alleles from Dongnong L13 proved to be favorable allelic genotypes for qPN-H-1 and qPN-L-4 QTL, whereas Henong 60 carried favorable allelic genotypes for the other 24 QTLs. To integrate these favorable alleles into one line, individuals carrying different combinations of favorable alleles which could be predicted by Bayesian probability, should be selected for crossing. The offspring could then be screened using marker-assisted selection.
Verification of QTLs
The authenticity of QTLs can be verified by a comparison of the genomic regions containing QTLs in different genetic backgrounds, or by identifying a single QTL that regulates multiple related traits.
To compare genome regions containing QTLs, we integrated all genomic fragments encompassing pod-number QTLs found in the present and previous studies into the pubic linkage map of the soybean genome [22] (S1 Table). Of all the genomic regions associated with the 47 QTLs identified in the present research, 14 and 21 regions associated with pod-number-related traits in RIL3613 and RIL6013 overlapped with those identified in previous reports, respectively.
A total of 11regions contained QTLs detected in both associated RIL populations (Table 2, S1 Table). The qPN-C1-1 region (24.11–41.43 cM; Satt396–Sat_140) overlapped the qPN-C1-2 region (28.04–41.43 cM; sat_367–sat_140) and the qPN-C2-1 region (107.58–112.34 cM; Satt277–Satt289) overlapped the qPN-C2-2 region (97.83–121.26 cM; satt376–satt307). The qPN-D1a-1 region (45.75~56.43 cM; satt482~satt254) overlapped the qPN-D1a-2 region (53.66–55.68 cM; Sat_346–Satt515). The qPN-D1b-3 region (102.59–112.62 cM; Sat_069–Sat_183), the qPN-D1b-1 region (0–131.91 cM; sat_096–sat_289), and the qPN-D1b-2 region (87.19–126.44 cM; satt546–staga002) all overlapped each other. The qPN-G-3 region (50.52–100 cM; satt352–sat_117) overlapped the qPN-G-1 region (62.08~68.76 cM; Sat_203~Satt503). The qPN-I-1 region (18.5–82.77 cM; Satt571–Satt292), the qPN-I-3 region (18.5–97.04 cM; Satt571–GMGLPSI2) and the qPN-I-2 region (27.98–77.83 cM; satt367–satt330) all overlapped each other. The qPN-L-1 region (33.7–78.23 cM; Satt497–Sat_099) overlapped the qPN-L-3 region (30.83–64.66 cM; sat_195–satt448) and the qPN-L-1 region (33.7–78.23 cM; Satt497–Sat_099) overlapped the qPN-L-4 region (34.54~107.23 cM; satt313~satt373) too. The qPN-O-1 region (14.17–118.13 cM; Satt500–Satt153), the qPN-O-2 region (28.95–51 cM; BF008905–Sat_221) overlapped the qPN-O-2 region (54.2–67.93 cM; Satt479–Sat_341). These 11 regions containing overlapping QTLs represent strong candidates for breeding programs to affect seed set and pod number.
Pleiotropic effects detected for a QTL also indicate its validity. Among the 47 QTLs detected in the present study, 23 were found to control multiple traits. In RIL3613, seven QTLs (qPN-A1-2, qPN-C1-1, qPN-D1a-2, qPN-D2-1, qPN-D2-2, qPN-I-3, and qPN-M-1) controlled two traits, one QTLs (qPN-B2-1) affected three traits. In RIL6013, five QTLs (qPN-B1-3, qPN-D1a-1, qPN-D1b-1, qPN-E-3 and qPN-G-3) controlled two traits, four QTLs (qPN-B1-2, qPN-D1b-2, qPN-G-2 and qPN-G-4) were associated with three traits, five QTLs (qPN-B1-1, qPN-B1-4, qPN-I-2, qPN-L-3 and qPN-O-2) regulated four traits, and one QTLs (qPN-N-1) conferred five traits. The association of these regions with multiple traits suggest the authenticity of these 23 QTLs.
Conclusion
A total of 21 and 26 QTLs controlling pod-number-related traits were identified in RIL3613 and RIL6013, respectively. Of all the identified QTLs, 32 QTLs were confirmed by comparison with previous research, were found to regulate multiple traits, or were identified in both associated RIL populations.
Supporting information
S1 Fig. The frequency distribution of pod-number-related traits.
https://doi.org/10.1371/journal.pone.0195830.s001
(DOCX)
S1 Table. Genomic region of QTLs associated with pod number-related traits detected in present and previous research.
https://doi.org/10.1371/journal.pone.0195830.s004
(DOCX)
S2 Table. Linkage map information for RIL3613 and RIL6013.
https://doi.org/10.1371/journal.pone.0195830.s005
(DOCX)
Acknowledgments
The authors gratefully acknowledge the financial support for this study provided by Natural Science Foundation of Heilongjiang Province [China] (LC201610) to H.N and, (C2015007) to W-X.L., and the Scientific and Technological Foundation of Education Department of Heilongjiang Province (12541z001) to H.N, grants from the National Key Research and Development Program of China (2017YFD0101300) to H.N.
References
- 1. Machado B, Nogueira A, Hamawaki O, Rezende G, Jorge G, Silveira I, et al. Phenotypic and genotypic correlations between soybean agronomic traits and path analysis. Genetics & Molecular Research, 2017, 16(2): gmr16029696. http://dx.doi.org/10.4238/gmr16029696
- 2. Vieira A J D, Oliveira D A, Soares T C B, Schuster I, Piovesan N, Martinez C, et al. Use of the QTL approach to the study of soybean trait relationships in two populations of recombinant inbred lines at the F7 and F8 generations. Brazilian Journal of Plant Physiology, 2006, 18(2): 281–290.
- 3. Sun D, Li W, Zhang Z, Chen Q, Ning H, Qiu L, et al. Quantitative trait loci analysis for the developmental behavior of soybean (Glycine max L. Merr.). Theoretical and Applied Genetics, 2006, 112(4): 665–673. pmid:16365761
- 4. Zhao G, Wang J, Han Y, et al. Identification of QTL underlying the resistance of soybean to pod borer, Leguminivoraglycinivorella, (Mats.) obraztsov, and correlations with plant, pod and seed traits. Euphytica, 2008, 164(1):275–282. https://doi.org/10.1007/s10681-008-9728-z
- 5. Li C, Jiang H, Zhang W, Qiu P, Liu C, Li W, et al. QTL analysis of seed and pod traits in soybean. Molecular Plant Breeding, 2008, 6(6):1091–1100.
- 6. Zhou R, Chen H F, Wang X Z, Zhang X J, Shan Z H, Wu X J, et al. QTL analysis of yield, yield components, and lodging in soybean. Acta Agronomica Sinica 2009, 35(5): 821−830
- 7. Palomeque L, Liu L-J, Li W, Hedges B, Cober E, Rajcan I. QTL in mega-environments: II. Agronomic trait QTL co-localized with seed yield QTL detected in a population derived from a cross of high-yielding adapted × high-yielding exotic soybean lines. Theoretical and Applied Genetics, 2009, 119(3):429–436. pmid:19462149
- 8. Yang Z, Xin D, Liu C, Jiang H, Han X, Sun Y, Qi Z, et al. Erratum to: Identification of QTLs for seed and pod traits in soybean and analysis for additive effects and epistatic effects of QTLs among multiple environments. Molecular Genetics and Genomics, 2013, 288(12):651. pmid:24022198
- 9. Kuroda Y, Kaga A, Tomooka N, Yano H, Takada Y., Kato S, et al. QTL affecting fitness of hybrids between wild and cultivated soybeans in experimental fields. Ecology & Evolution, 2013, 3(7):2150–68. pmid:23919159
- 10. Rodrigues J, Miranda F, Piovesan N, Ferreira A, Ferreira M, Cruz C, et al. QTL mapping for yield components and agronomic traits in a Brazilian soybean population. Crop Breed. appl. biotechnol, 2016, 16(4):265–273.
- 11. Liu N, Li M, Hu X, Ma Q, Mu Y, Tan Z, et al. Construction of high-density genetic map and QTL mapping of yield-related and two quality traits in soybean RILs population by RAD-sequencing. Bmc Genomics, 2017, 18(1):466. pmid:28629322
- 12. Li D, Sun M, Han Y, Teng W, Li W. Identification of QTL underlying soluble pigment content in soybean stems related to resistance to soybean white mold (Sclerotiniasclerotiorum). Euphytica, 2010, 172(1):49–57. https://doi.org/10.1007/s10681-009-0036-z
- 13. Zhang D, Cheng H, Wang H, Zhang H, Liu C, Yu D. Identification of genomic regions determining flower and pod numbers development in soybean (Glycine max L.). Journal of Genetics and Genomics, 2010, 37(8): 545–556. pmid:20816387
- 14. Wang X, Zhang X, Zhou R, Sha A, Wu X, Cai S, et al. QTL analysis of seed and pod traits in soybean RIL population. Acta Agronomica Sinica 2007, 33(3):441–448.
- 15. Chen Q, Zhang Z, Liu C, Xin D, Qiu H, Shan D, et al. QTL analysis of major agronomic traits in soybean. Agricultural Sciences in China, 2007, 6(4): 399–405.
- 16. Liu W, Kim M, Van K, Lee Y, Li H., Liu X, et al. QTL identification of yield-related traits and their association with flowering and maturity in soybean. Journal of Crop Science and Biotechnology, 2011, 14(1): 65–70.
- 17. Wu X, Wand Y, He C, Chen S, Gai J, Wang X. QTLs mapping of some agronomic traits of soybean. Acta Genetica Sinica, 28(10):947–955, 2001 pmid:11695267
- 18. Bobby R, Bazelle R, Clark W, Kantartzi S K, Meksem K, Akond M, et al. Genetic analysis of yield components in the PI 438489B by 'Hamilton' Recombinant Inbred Line (RIL) Population of Soybean [Glycine max (L.) Merr.]. Journal of Agricultural Science, 2012, 4(9): 98–105.
- 19. Liu B, Qu D, Zhou X. The shoot dry matter accumulation and vertical distribution of soybean yield or yield components in response to light enrichment and shading. Emirates Journal of Food & Agriculture, 2015, 27(3): 258–265.
- 20. Liu Y, Li R, Li Z. Effects of cultivation pattern and density on canopy structure and yield of soybean. Soybean Science, 2010, 29(5):796–795.
- 21. Doyle J. Isolation of plant DNA from fresh tissue. Focus, 1990,12:13–15.
- 22. Song Q, Marek L, Shoemaker R, Lark K, Concibido V, Delannay X, et al. A new integrated genetic linkage map of the soybean. Theoretical and Applied Genetics, 2004, 109(1):122–8. pmid:14991109
- 23. Wang J. Inclusive composite interval mapping of quantitative trait genes. Acta Genetica Sinica, 2009, 35(2): 239−245.
- 24. Zeng Z.-B. (1994) Precision mapping of quantitative trait loci. Genetics 136: 1457–1468. pmid:8013918