Abstract:

The objective of this study was to evaluate genomic prediction using a breed origin of alleles (BOA) model for a multi-breed Irish beef cattle population. Our BOA model uses one matrix with allele counts for each breed to predict breed-specific marker effects, enabling to use crossbred and purebred animals in the reference population. Pedigree-based best linear unbiased prediction (PBLUP), a genomic model assuming a homogeneous population (SNP-BLUP) and BOA model were compared. Accuracy was estimated as the correlation between breeding values and corrected phenotype divided by the square root of heritability. The accuracy of predictions using BOA increased by 82 and 42% for crossbred and purebred animals, respectively, compared with PBLUP, and improvements of BOA over SNP-BLUP were of 6% for crossbred animals.