Next-Generation Soybean Haplotype Map as A Genomic Resource for Enhanced Trait Discovery and Functional Analysis

Abstract We present a global soybean haplotype map generated from whole-genome sequencing of 1,278 Glycine max and Glycine soja accessions, comprising 11.37 million SNPs and 2.05 million short insertions and deletions. This map (GmHapMap-II) captures unprecedented worldwide genetic diversity, reflecting the broad extent of the global soybean gene pool. Population structure analyses revealed six geographically distinct subpopulations that affected the linkage and shaped the recombination. The haplotype variation map was used to identify novel genomic regions associated with crude protein content on chromosome 15 that were not detected by a lower SNP density array. LD-based haplotype analysis revealed a superior haplotype for crude protein content. The constructed haplotype map enabled detailed characterization of haplotype diversity and copy number polymorphism at the SCN-associated rhg-1 and Rhg-4 loci, revealing both novel haplotype structures and germplasm lines with elevated CNV relative to previously characterized genotypes. We employed the HapMap matrix for a multi-class variations ML-based genomic prediction approach to predict phenotypes for SCN and catalogued the gene-centric haplotypes in a user-friendly database. The analysis revealed the extent of deleterious alleles present in the soybean germplasm and how breeders have deployed beneficial alleles and purged deleterious ones. The haplotype map will serve as a major genomic resource for trait-based mapping, enhancing efforts in the genomics-enabled development of improved cultivars.