Bayesian genomic models boost prediction accuracy for resistance against Streptococcus agalactiae in Nile tilapia (Oreochromus nilioticus)

Abstract

Streptococcosis due to Streptococcus agalactiae is a major bacterial disease in Nile tilapia, and development of the resistant genetic strains can be a sustainable approach towards combating this problematic disease. Thus, a controlled disease trial was performed on 120 full-sib families to i) quantify and characterize the potential of genomic selection for S. agalactiae resistance in Nile tilapia and to ii) select the best genomic model and optimal SNP-chip for this trait. In total, 40 fish per family (15 fish intraperitoneally injected and 25 fish as cohabitants) were selected for the challenge test and mortalities recorded every 3 hours, until no mortalities occurred for a period of 3 consecutive days. Genotypes (50,690 SNPs) and phenotypes (0 for dead and 1 for alive) for 2472 cohabitant fish were available. The pedigree-based analysis utilized a deep pedigree, going 17 generations back in time. Genetic parameters were obtained using various genomic selection models (GBLUP, BayesB, BayesC, BayesR and BayesS) and traditional pedigree-based model (PBLUP). The genomic models were further analyzed using 10 different subsets of SNP-densities for optimum marker density selection. Prediction accuracy and bias were evaluated using 5 replicates of 10-fold cross-validation. Using an appropriate Bayesian genomic selection model and optimising it for SNP density increased prediction accuracy up to [~]71%, compared to a pedigree-based model. This result is encouraging for practical implementation of genomic selection for S. agalactiae resistance in Nile tilapia breeding programs.