Abstract
Grass carp (Ctenopharyngodon idella), a globally important aquaculture species, exhibits protein-dependent growth plasticity, requiring genetic improvement for sustainable production. This study integrates genome-wide association analysis (GWAS) and genomic selection (GS) to unravel the genetic architecture of four growth traits, body weight, body length, body height, and body depth, in grass carp fed with diets of varying protein levels (20%, 25%, 30%, and 35%). Using a 21K liquid SNP array, we identified 62,736 high-quality SNPs across 24 chromosomes, with 90 SNPs significantly associated with growth traits. Notably, three SNPs (SLG14_24417024, SLG14_24417039, SLG24_30276273) exhibited pleiotropic effects on multiple traits. Functional annotation of 276 candidate genes near significant SNPs revealed enrichment in keratinocyte development, septin cytoskeleton organization, and heat acclimation pathways. Genomic prediction achieved accuracies up to 0.79 for body weight traits using 1533 optimal markers. This study provides the first comprehensive SNP resource for grass carp growth traits with different dietary treatments, bridging GWAS and genomic prediction to accelerate marker-assisted selection. Our findings not only advance genetic breeding strategies but also inform protein diet optimization, minimizing economic and environmental costs in aquaculture.