Machine Learning based Genome-Wide Association Studies for Uncovering QTL Underlying Soybean Yield and its Components

Abstract

Genome-wide association study (GWAS) is currently one of the important approaches for discovering quantitative trait loci (QTL) associated with traits of interest. However, insufficient statistical power is the limiting factor in current conventional GWAS methods for characterizing quantitative traits, especially in narrow genetic bases plants such as soybean. In this study, we evaluated the potential use of machine learning (ML) algorithms such as support vector machine (SVR) and random forest (RF) in GWAS, compared with two conventional methods of mixed linear models (MLM) and fixed and random model circulating probability unification (FarmCPU), for identifying QTL associated with soybean yield components. In this study, important soybean yield component traits, including the number of reproductive nodes (RNP), non-reproductive nodes (NRNP), total nodes (NP), and total pods (PP) per plant along with yield and maturity were assessed using 227 soybean genotypes evaluated across four environments. Our results indicated SVR-mediated GWAS outperformed RF, MLM and FarmCPU in discovering the most relevant QTL associated with the traits, supported by the functional annotation of candidate gene analyses. This study for the first time demonstrated the potential benefit of using sophisticated mathematical approaches such as ML algorithms in GWAS for identifying QTL suitable for genomic-based breeding programs.