Data_Sheet_1_Genotyping of Soybean Cultivars With Medium-Density Array Reveals the Population Structure and QTNs Underlying Maturity and Seed Traits.ZIP

Soybean was domesticated about 5,000 to 6,000 years ago in China. Although genotyping technologies such as genotyping by sequencing (GBS) and high-density array are available, it is convenient and economical to genotype cultivars or populations using medium-density SNP array in genetic study as well as in molecular breeding. In this study, 235 cultivars, collected from China, Japan, USA, Canada and some other countries, were genotyped using SoySNP8k iSelect BeadChip with 7,189 single nucleotide polymorphisms (SNPs). In total, 4,471 polymorphic SNP markers were used to analyze population structure and perform genome-wide association study (GWAS). The most likely K value was 7, indicating this population can be divided into 7 subpopulations, which is well in accordance with the geographic origins of cultivars or accession studied. The LD decay rate was estimated at 184 kb, where r2 dropped to half of its maximum value (0.205). GWAS using FarmCPU detected a stable quantitative trait nucleotide (QTN) for hilum color and seed color, which is consistent with the known loci or genes. Although no universal QTNs for flowering time and maturity were identified across all environments, a total of 30 consistent QTNs were detected for flowering time (R1) or maturity (R7 and R8) on 16 chromosomes, most of them were corresponding to known E1 to E4 genes or QTL region reported in SoyBase (soybase.org). Of 16 consistent QTNs for protein and oil contents, 11 QTNs were detected having antagonistic effects on protein and oil content, while 4 QTNs soly for oil content, and one QTN soly for protein content. The information gained in this study demonstrated that the usefulness of the medium-density SNP array in genotyping for genetic study and molecular breeding.

大豆（Soybean）约于5000至6000年前在中国被驯化。尽管目前已有测序分型（genotyping by sequencing, GBS）及高密度芯片等基因分型技术，但在遗传研究与分子育种中，采用中等密度单核苷酸多态性（single nucleotide polymorphisms, SNP）芯片对栽培品种或群体进行基因分型，依然具备便捷且经济的优势。本研究利用包含7189个单核苷酸多态性（SNP）的SoySNP8k iSelect微珠芯片，对采自中国、日本、美国、加拿大及其他国家的235份栽培品种进行基因分型。最终筛选得到4471个多态性SNP标记，用于群体结构分析及全基因组关联分析（genome-wide association study, GWAS）。群体结构分析显示最适K值为7，表明该供试群体可划分为7个亚群，这一结果与供试品种或材料的地理起源高度吻合。本研究估算的连锁不平衡（linkage disequilibrium, LD）衰减速率为184 kb，此时r²降至最大值（0.205）的一半。采用FarmCPU模型进行全基因组关联分析，检测到与种脐颜色和籽粒颜色相关的稳定数量性状核苷酸（quantitative trait nucleotide, QTN），该结果与已知位点或基因相一致。尽管未在所有环境条件下检测到通用的开花期与成熟期相关数量性状核苷酸，但在16条染色体上共检测到30个与开花期（R1）或成熟期（R7、R8）相关的一致性数量性状核苷酸，其中多数对应大豆数据库（SoyBase, soybase.org）已报道的E1至E4基因或数量性状位点区域。在16个与蛋白质及油脂含量相关的一致性数量性状核苷酸中，11个数量性状核苷酸对蛋白质和油脂含量存在拮抗效应，4个仅与油脂含量相关，1个仅与蛋白质含量相关。本研究获得的结果证实，中等密度SNP芯片在用于遗传研究与分子育种的基因分型工作中具备实用价值。