Comprehensive Survey for Genes Contributing to Soybean Domestication and Improvement

Soybean is a crop with substantial economic value, accounting for more than half of global oilseed production. It has been suggested that cultivated soybean was domesticated from wild soybean in China 5,000 years ago. Comprehensive evaluation of large scale of representative germplasm and survey for genes contributing to domestication and improvement are essential for the super variety breeding.

 

A collaborate team led by Dr. TIAN Zhixi from Institute of Genetics and Developmental Biology, the Chinese Academy of Sciences and Dr. WANG Wen from Kunming Institute of Zoology, the Chinese Academy of Sciences, performed a large scale assessment of soybean domestication and improvement by resequencing of 302 wild, landrace and cultivated soybean lines with an average coverage depth of more than 11× for each line.

 

Bioinformatics analysis identified a total of 121 and 109 selective sweeps during soybean domestication and improvement, respectively. Genome-wide association study revealed associations between 10 selected regions and 9 domestication or improvement traits and identify 13 previously uncharacterized loci for agronomic traits including oil content, plant height and pubescence form. Further investigation by combining previous quantitative trait loci (QTL) information revealed that 96 of the 230 selected regions correlate with reported oil QTLs and 21 contained fatty acid biosynthesis genes.

 

This study provides a resource to improve our understanding of the genetics of soybean domestication and to inform future studies on the allelic variation of relevant traits within genetic resource collections, thereby enabling soybean crop improvement.

 

This work entitled “Resequencing 302 wild and cultivated accessions identifies genes related to domestication and improvement in soybean” was published in Nature Biotechnology online Early Edition on 2 February 2015 (DOI:10.1038/nbt.3096), with Dr. ZHOU Zhengkui, Dr. WANG Zheng, Dr. LI Weiyu from Dr. TIAN Zhixi’s lab and Dr. JIANG Yu, Dr. LYU Jun, and Mr. GOU Zhiheng from Dr. WANG Wen’s lab as co-first authors.

 

This research was supported by grants from National Basic Research Program of China, National Natural Science Foundation of China, and Strategic Priority Research Program of the Chinese Academy of Sciences.

 

 

Figure 1. Geographic distribution and population structure of 302 soybean accessions.

 

 (a) The geographic distribution of the 302 accessions, each of which is represented by a dot on the world map. (b) Phylogenetic tree of all accessions inferred from whole-genome SNPs, with Medicago truncatula as an outgroup. The layer rings indicate the group name of each clade. (c) PCA plots of the first two components of 302 accessions. (d) The geographic origin of each accession in the eight clades. J&K, Japan and Korea; SC, southern China; NC, northern China; NEC, northeastern China; NUS, northern United States; SUS, southern United States; CAN, Canada; Other, other country.  (Image by IGDB)

 

 

 

Figure 2. Genome-wide screening and functional annotations of selected regions during domestication and improvement. (a) and (f) show the whole-genome screening of selective signals in domestication and improvement, respectively. The XP-CLR values are plotted against the position on each of the 20 chromosomes. The horizontal dashed lines indicate the genome-wide threshold of selection signals. For domestication (a), the threshold is XP-CLR ≥ 38, and for improvement (f), the threshold is XP-CLR ≥ 4.6. The previously reported QTL regions21, which mapped 9 domestication-related traits in soybean, are marked with yellow bars above the x-axes. Four known causal genes that overlapped selective signals-E1, Sg1, E2 and W1 are labeled in red font, and the E1 signal is moderately lower than the threshold. Selection signals that overlap soybean oil QTLs are labeled with turquoise bars, and the QTL names assigned by SoyBase45 (http://www.soybase.org/) are marked above. Selective sweeps overlapping characterized GWAS loci and known causal genes are shown in dark green. b-e, g and h show six GWAS results that overlapped strong selective signals. (Image by IGDB)