Genomic selection for hypoxia tolerance in large yellow croaker

The large yellow croaker (Larimichthys crocea) is one of the most economically important marine fishes on the southeast coast of China. However, high-density cultures, inbreeding depression, and environmental pollution have resulted in L. crocea often experiencing hypoxia, which has caused huge economic losses to the L. crocea farming industry. This study aimed to improve the hypoxia tolerance of farmed L. crocea using genomic selection (GS) for breeding. In this study, 399 and 354 samples from the reference and candidate populations, respectively, were genotyped using a 55 K liquid single-nucleotide polymorphism (SNP) array. Based on two different representations (survival status and survival time), the estimated values of heritability for the hypoxia tolerance trait in L. crocea using 38,472 clean SNPs were 0.65 and 0.62, respectively. A comparison of the genomic prediction accuracy under different GS models and SNP densities showed that GBLUP was the best GS model, and the 2500 SNPs selected using a genome-wide association study (GWAS)-based strategy were essentially close to all 38,472 SNPs. Based on the established optimal GS strategy (prediction accuracy was 0.4901 ± 0.0246), genomic estimated breeding values (GEBVs) were calculated for each candidate population. The tolerance capacity tests showed a significant improvement in hypoxia tolerance in the offspring of the candidate population with the top 10% GEBVs, having an 13.40% decrease and a 34.02% increase in the median lethal concentration (LC50) and median lethal time (LT50), respectively, compared to the control group. These results confirmed that GS is an effective genetic breeding technique when selecting for hypoxia tolerance in L. crocea.