Soybean is one of the most valuable and profitable oil crop species and a thorough knowledge of the genetic structure of this crop is necessary for developing the best breeding strategies. In this study, a representative collection of soybean cultivars recommended for farming in all Brazilian regions was genotyped using 27 simple sequence repeat (SSR) loci. A total of 130 alleles were detected, with an average allelic number of 4.81 per locus. These alleles determined the core set that best represented this soybean germplasm. The Bayesian analysis revealed the presence of two clusters or subgroups within the whole collection (435 soybean cultivars) and the core set (31 entries). Cultivars of similar origin (ancestral) were clustered into the same groups in both analyses. The genetic diversity parameters, based on the SSR loci, revealed high similarity between the whole collection and core set. Differences between the two clusters detected in the core set were attributed more to the frequency of their ancestors than to their genetic base. In terms of ancestry, divergent groups were presented and a panel is shown which may foster efficient breeding programs and aid soybean breeders in planning reliable crossings in the development of new varieties.
Keywords: Bayesian method; Glycine max (L.) Merrill; SSR markers; core collection; genetic base.