Genomic selection accelerates genetic improvement of tolerance to visceral nodular disease in the half-smooth tongue sole (Cynoglossus semilaevis).

The half-smooth tongue sole (Cynoglossus semilaevis) is an economically important marine fish in China, but its vulnerability to visceral nodular disease (VND) caused by Mycobacterium marinum poses a significant threat to aquaculture. Based on traditional breeding methods, it is slowly to improve disease tolerance and then may lead to inbreeding depression. This study aimed to accelerate the development of VND-tolerant strains of C. semilaevis through genomic selection (GS). In this study, individuals from both reference and candidate populations were genotyped, and more than 7 million high-quality SNPs were identified by whole genome sequencing (WGS). Heritability estimates for VND tolerance, based on binary survival status and survival time, were 0.37 and 0.39, respectively, indicating moderate genetic control. To improve breeding efficiency, six GS models (Bayes A, Bayes B, BayesCpi, ssGBLUP, GBLUP, and rrBLUP) were evaluated, of which GBLUP showed the highest prediction accuracy (0.4925 for binary state and 0.5477 for survival time). Furthermore,the effect of SNP density on genomic prediction accuracy was investigated and it was found that 40K SNPs, selected via GWAS, provided the most reliable predictions. Through the GBLUP model with 40K SNPs, genomic estimated breeding values (GEBVs) were estimated for the candidate population and top 20 % was selected for breeding. Progeny from these selected individuals showed significantly higher survival rates in response to VND challenge, with 19.7 % increase compared to the control group, and these results confirmed the effectiveness of GS in enhancing disease tolerance. This study provides a promising approach to developing VND-tolerant C. semilaevis, which could significantly improve the sustainability of aquaculture.