Comparative Study of Single-Trait and Multi-Trait Genomic Prediction Models.

ABSTRACT: Conventional genomic selection models trait individually, neglecting complex trait interactions. Multi-trait models address this by considering genetic correlations, thus improving breeding value accuracy. Despite their theoretical benefits, quantifying these models' breeding advantages across genetic backgrounds is essential. This study evaluates the benefits of multi-trait models under varying population sizes and three levels of genetic correlations (low, medium, high) using simulations based on 50 K chip data from 5000 individuals. In equal heritability scenarios, the multi-trait GBLUP model consistently outperforms single-trait models, with breeding advantages increasing with heritability. For example, with a reference population of 4500, improvements range from 0.3% to 4.1%. Notably,

SUBMITTER: Tang X 

PROVIDER: S-EPMC11506144 | biostudies-literature | 2024 Oct

REPOSITORIES: biostudies-literature