Abstract

Nowadays, studies on genetic by environment interactions (GxE) are receiving increasing attention because of its theoretical and practical importance in explaining individual behavioral traits. Use information from different environments to improve the statistical power of genome-wide association and prediction in the hope of obtaining individuals with better breeding value is the most expedient way. However, there are significant challenges when performing genome-wide association studies (GWAS) and genomic selection (GS) using multiple environments or traits, mainly because most diseases and quantitative traits have numerous associated loci with minimal effects. Therefore, this study constructed a new genotype design model program (GbyE) for genome-wide association and prediction using Kronecker product, which can enhance the statistical power of GWAS and GS by utilizing the interaction effects of multiple environments or traits. The data of 282 maize, 354 yaks and 255 peaches were used to evaluate the power of the model at different levels of heritability and genetic correlation. The results show that GbyE can provide higher statistical power for the traditional GWAS and GS models in any heritability and genetic correlation, and can detect more real loci. In addition, GbyE has increased statistical power to three Bayesian models (BRR, BayesA, and BayesCpi). GbyE can make full use of multiple environment or trait informations to increase the statistical power of the model, which can help us understand the GxE and provide a method for predicting association loci for complex traits.