Network analysis identifies QTLs and relationships between seed protein and free amino acids in soybean

Abstract Protein and amino acid profile in soybean seeds are important factors for its market price, quality of the animal feed produced and producer’s income. With a soybean panel of 189 plant introductions, a genome-wide association study (GWAS) was performed for protein and free amino acid (FAA) content (leaf vs. seed). Correlation analysis of FAA and protein contents identified strong relationships between protein and asparagine (Asn), glutamine relative content in seeds, but not with leaf-FAAs. Significant protein-associated linkage disequilibrium (LD) blocks were identified only on Chromosome 20 known for genomic regions controlling protein/FAA contents along with two novel quantitative trait loci (QTL) on the same chromosome and their significance levels were quite different depending on each environment. For FAA contents, six significant LD blocks for aspartate, lysine, and valine among leaf-FAAs were identified, whereas twelve LD blocks related to Asn, γ-aminobutyric acid (GABA), histidine, methionine, phenylalanine, serine, and threonine were detected by GWAS conducted with seed FAA contents. Five and seven novel QTL for leaf- and seed-FAAs, respectively, were identified. GWAS combined with a correlation-based network analysis utilized network-derived FAA traits from leaf and seed. Seven novel QTL for network-derived traits were detected with leaf-FAAs and three novel QTL for absolute seed-FAAs were also identified. Comparing the candidate genes within the significant genomic regions provided insight into how FAAs are involved in seed development and sink-source partitioning. Overall, network analysis can help to identify genes putatively regulating AA content in soybean seeds, contributing to the development of new and improved cultivars.