Abstract
In order to investigate the applicability and efficiency of genomic selection for growth and carcass traits in Nordic beef cattle, single-step genomic BLUP (ssGBLUP) was applied in 4321 Charolais and 4532 Hereford animals with information on approximately 43,000 SNPs each. Statistics including dispersion value (b1), accuracy ratio and the relative accuracy improvement were estimated for genotyped female animals in the validation set. For estimating dispersion, accuracy ratio and relative accuracy improvement, the Legarra-Reverter linear regression (LR) method was used by truncating the phenotypes after 2018, and the validation set comprised females born from 2019 to 2021. Moreover, for ssGBLUP, different alpha values of 0.95 and 0.70 were utilised as weights on the genomic information when the H matrix was blended for the genomic relationship matrix G and the pedigree relationship matrix A. In general, implementing ssGBLUP led to higher accuracy ratios and improved dispersion values (b1 value closer to the optimum value of one), compared to when using pedigree-based BLUP (PBLUP). Using an alpha value of 0.70 gave a dispersion value closer to one compared with when using an alpha value of 0.95. Additionally, the relative accuracy estimation was improved substantially for several traits by using ssGBLUP instead of PBLUP, with the highest (30%) relative improvement for carcass conformation in Swedish Hereford cattle. In conclusion, ssGBLUP would be beneficial to implement in the future Nordic beef cattle breeding programs.