287. Effect of informative polymorphism on accuracy of genomic prediction and heritability

Abstract

Early fertility traits are linked with beef cattle profitability, and so they have become target traits for selective breeding. Informative single nucleotide polymorphism (SNP) can potentially increase the accuracy of genomic predictions for target traits. Multi-trait meta-analysis was applied on two reproductive traits measured early in life: pregnancy outcome after the first mating opportunity (PREG1) and rebreeding within the first two mating opportunities (REB) to identify informative SNP associated with these traits. The accuracy of genomic prediction using informative SNP was compared to using the bovine high density (HD) SNP chip in full. Accuracies were calculated in five-fold cross validations, through genomic best linear unbiased prediction (GBLUP) methods. The accuracy of genomic prediction for PREG1 was similar for both SNP sets: 0.18 for the informative SNP set, 0.19 for the HD SNP set. However, the accuracy of genomic prediction for REB was higher when the informative SNP set was used (0.36), in comparison to the HD SNP set (0.19).