Understanding University of Guelph's Soybean Germplasm Diversity Using GBS-SNP and SSR Data.

Combining traditional breeding approaches with molecular tools will lead to an improved understanding and exploitation of genetic diversity within a breeding program that may result in the development of new superior soybean cultivars. A panel of ninety-six cultivars from the University of Guelph’s soybean breeding program comprised of current cultivars, experimental lines, pedigree ancestors and diverse-cross derived recombinant inbred lines (RILs) were genotyped using genotyping-by-sequencing (GBS) single nucleotide polymorphism (SNP) and simple sequence repeat (SSR) approaches. The filtered GBS-SNP dataset provided roughly 25x the marker density of the SSR dataset. The two genotyping methods were compared for their ability to cluster breeding germplasm, identify haplotype blocks within the germplasm, assess genome-wide diversity, and identify genomic regions under selective pressure. Modern lines were compared to their pedigree ancestors to determine the reduction in diversity as a direct result of breeders’ selections over 100 years of soybean improvement. A test example using this data is illustrated in the analysis of anthocyanin pathway genomic diversity across cycles of selection. As this pathway is under high selective pressure by breeders due to seed and hilum color phenotypes, it is expected that the two genotyping systems will demonstrate the effects of selection at loci within the anthocyanin pathway. The data will also be used for genome-wide association analysis of agronomic traits including seed yield, for the understanding of transmission of alleles that condition these traits through long-term selection. It is expected that this research will lead to an improved understanding of the use of GBS-SNP and SSR data in a breeding context, and the differences in the practical application between the two systems. The long term goal of this research is to facilitate improved breeder selections by combining traditional breeding approaches with the genomic data.