Potential of genome-wide association studies to improve genomic selection for growth traits in blotched snakehead (Channa maculata)

Abstract

Listen

Translate article

Blotched snakehead (Channa maculata) holds significant economic importance in China as a freshwater fish species, with its growth trait serving as a cornerstone in selective breeding programs. Genomic selection (GS) has been proven to be an effective breeding method that can significantly accelerate breeding processes. In this study, 790 blotched snakeheads were genotyped using a 50 K SNP array, resulting in the identification of 45,695 high-quality SNPs. Based on these genome-wide SNPs, we assessed population structure and relatedness among the samples, and calculated the heritability estimates for weight and total length traits as 0.31 ± 0.07 and 0.29 ± 0.07, respectively. We compared the predictive performance of four models—GBLUP, BayesA, BayesB, and BayesC—and found that the GBLUP model exhibited varying predictive power for the two growth traits. Additionally, we evaluated different SNP selection strategies and found that GWAS-selected SNPs significantly improved predictive accuracy compared to using all SNPs, with enhancements of 35.90 % for weight and 26.67 % for total length. Remarkably, optimal prediction accuracy for weight traits could be attained with GWAS-selected SNP panels as modest as 0.5 K. Moreover, when employing random SNP panels, prediction accuracy for weight stabilized at 1 K SNPs, while for total length, stabilization occurred at 3 K SNPs. These findings underscore not only the viability of employing low-density SNP panels in GS for blotched snakehead but also highlight the promising application prospects of integrating GWAS with GS in selective breeding practices. This study provides crucial insights for GS of growth traits in blotched snakehead, offering a pivotal reference for expediting the application of GS within aquaculture species.