Genomic Selection and Genome-Wide Association Analysis for Stress Response, Disease Resistance and Body Weight in European Seabass

SIMPLE SUMMARY: In the present study, the genetic architecture of the stress response, body weight, and disease resistance in European seabass were studied, providing fruitful results for further research. Initially, the above traits were recorded and genotyping on a large scale was performed in those fish. The recorded data and genotypes were combined and analyzed to find genomic areas affecting them as well as to estimate the heritability of those traits. Stress response traits and body weight were medium heritable, while genomic regions affecting them were detected. However, no genomic areas related to disease resistance were revealed. These findings improve our knowledge of the genetic structure of those traits and can be utilized in a breeding program for the genetic improvement of aquaculture broodstocks. ABSTRACT: The majority of the genetic studies in aquaculture breeding programs focus on commercial traits such as body weight, morphology, and resistance against diseases. However, studying stress response in European seabass may contribute to the understanding of the genetic component of stress and its future use to select broodstock whose offspring may potentially be less affected by handling. A total of 865 European seabass offspring were used to measure body weight and stress response. Moreover, a disease challenge experiment with Vibrio anguillarum was conducted in a subset (332) of the above fish to study disease resistance. Fish were genotyped with a 57k SNP array, and a Genome-Wide Association study (GWAS) was performed. Five SNPs were found to be statistically significant, three of which affect stress indicators and body weight (in a subgroup of the population), and a putative SNP affects growth performance, while no SNP associated with resistance to Vibrio was found. A moderate to high genomic heritability regarding stress indicators and body weight was estimated using the Restricted Maximum Likelihood (REML) process. Finally, the accuracy, along with the correlation between Estimated Breeding Values (EBVs) and Genomic Estimated Breeding Values (GEBVs), were calculated for all the traits.