Accuracies of Genomic Breeding Values for Carcass Traits in Korean Hanwoo Cattle

The genetic improvement for economic traits in Hanwoo beef cattle has, like most livestock species, been achieved through breeding programs that rank animals based on phenotypic and pedigree information. Recently, genomic prediction using molecular markers has been a major research topic in order to obtain more reliable genetic evaluations. The objective of this study was to estimate accuracies of genomic breeding values using Illumina BovineSNP50 genotypes for carcass traits (carcass weight, eye muscle area, back-fat thickness and marbling score) in Korean Hanwoo beef cattle using 10-fold cross-validation. A total of 489 genotyped cattle were allocated to 10-groups using either K-means or random clustering methods for cross-validation. One group was used for validation while the other 9 groups were used for training, and this was repeated with each group used once for validation.  Accuracies of genomic breeding values were calculated as simple correlations between direct genomic values and phenotypic data adjusted for the two fixed effects (contemporary groups and age). The accuracies of DGV using the K-means (or random) clustering were 0.13 (0.10), 0.17 (0.37), 0.18 (0.29) and 0.16 (0.27) for the studied traits, respectively. These results established the basis for application of genomic prediction in Hanwoo beef cattle although the accuracies in this study are smaller than other published results for various beef cattle breeds. In order to improve the accuracy of DGV in Hanwoo beef cattle, continued research is needed in addition to an increase in the size of the training population.