Genetic Architecture of Growth Traits in Turbot (Scophthalmus maximus) Revealed by Genome-Wide Association Studies

Turbot (Scophthalmus maximus) is a key farmed flatfish species in North China, where growth performance directly determines aquaculture profitability. To elucidate the genetic architecture underlying growth traits, we performed whole-genome resequencing of 177 turbot individuals and obtained 5.29 million high-quality single nucleotide polymorphisms (SNPs). Genome-wide association studies (GWAS) were conducted for eleven growth traits—body weight (BW), total length (TL), body length (BL), body depth (BD), trunk length (TUL), head length (HL), snout length (SnL), caudal peduncle depth (CPD), eye diameter (ED), post-orbital head length (PoL), and interorbital width (IW)—using both single-trait and multi-trait approaches. Heritability estimates ranged from 0.001 (IW) to 0.259 (BW), with BW showing the highest heritability, and genetic correlation analysis revealed strong positive correlations between BW and most body size traits (rg > 0.7). Critically, extensive genetic pleiotropy governing these traits was uncovered. In single-trait GWAS, two pleiotropic loci—19:16017254 (within fam81b) and 15:2921091 (within atrn)—were repeatedly associated with the same six body-size traits (BD, BL, CPD, HL, TL, and TUL), demonstrating that single variants can exert coordinated effects across multiple morphological dimensions. Furthermore, multi-trait GWAS captured additional pleiotropic signals that remained undetected in single-trait analyses, identifying 20 novel loci encompassing key candidates such as iqgap1 (cytoskeletal signaling), enah (actin organization), and baz2a (chromatin remodeling). KEGG enrichment further consolidated these findings, with candidate genes significantly enriched in the “Regulation of actin cytoskeleton” pathway, while amd1 emerged as a hub gene integrating three metabolic pathways. Collectively, these results highlight the pervasive pleiotropy underlying turbot growth and serve as promising candidate loci for marker-assisted selection, although independent validation in larger populations is required before practical application.