Genetic diversity and yield trait associations in soybean germplasm using SSR markers

Soybean (Glycine max (L.) Merrill), a vital food and oilseed crop native to China, has the development of high-yield and high-quality new varieties as a core objective in soybean breeding. However, studies on genetic diversity and marker-trait associations for yield-related traits across multiple environments using simple sequence repeat (SSR) markers remain limited. This study utilized 20 pairs of SSR primers to investigate genetic diversity and marker-trait associations (MTAs) between SSR markers and yield-related traits in 150 soybean accessions of diverse origins, which were grown under two environmental conditions. Genetic diversity analysis of 12 key agronomic traits, including plant height, number of branches, number of pods per plant, and hundred-seed weight, showed substantial phenotypic variation with large ranges and high coefficients of variation among the accessions. Cluster analysis based on agronomic traits classified the germplasm into five clusters under environment E1 and four clusters under environment E2. SSR primer amplification of the accessions' genomic DNA produced an average of 7.10 amplified loci and 6.80 polymorphic loci per primer pair, with an average polymorphic locus rate of 96%. The average values of effective allele number (Ne), gene diversity index (H), Shannon's information index (I), polymorphism information content (PIC), and genetic similarity coefficient (GSC) were 1.5035, 0.3581, 0.5176, 0.3539, and 0.57, respectively. SSR-based cluster analysis divided the 150 accessions into three distinct clusters. Principal coordinate analysis (PCoA) validated that the selected SSR primers could effectively discriminate among the accessions. Population structure analysis indicated complex genetic relationships and substantial genetic diversity within the germplasm panel. Association analysis between agronomic traits and SSR markers identified 44 markers significantly associated with yield-related traits, of which 13 were simultaneously linked to two or more yield-related traits. Using generalized linear model (GLM) and mixed linear model (MLM) analyses, six SSR markers significantly associated with yield-related traits were consistently detected across the two environments. These results provide a theoretical basis for germplasm utilization and molecular breeding in soybean.