Genomic-polygenic and polygenic predictions for nine ultrasound and carcass traits in Angus-Brahman multibreed cattle using three sets of genotypes

The objectives of this study were to estimate variance components, genetic parameters, EBV, accuracies, and rankings for nine ultrasound and carcass traits in a multibreed Angus-Brahman population using three genomic-polygenic models and one polygenic model (PM). The genomic-polygenic models used the complete GeneSeek GPF250k SNP set (GPM), top 5% SNP (GPMR1), and 5% SNP evenly spread across the genome (GPMR2). Yearling ultrasound traits were weight (UW), ribeye area (UREA), backfat (UFAT), and percent intramuscular fat (UPIMF). Carcass traits were slaughter age (SLA), hot carcass weight (HCW), ribeye area (REA), backfat thickness (FAT), and marbling score (MAR). The 9-trait GPM, GPMR1, GPMR2, and PM contained fixed contemporary group, age of calf (ultrasound traits only), sex of calf, and direct heterosis effects, and random animal and residual effects. Variance components and genetic parameters were computed using AIREMLF90. Comparable heritabilities were obtained with GPM and PM for UW (GPM: 0.54 ± 0.05; PM: 0.51 ± 0.05), UREA (GPM: 0.36 ± 0.03; PM: 0.34 ± 0.03), UFAT (GPM: 0.12 ± 0.02; PM: 0.11 ± 0.02), UMPIMF (GPM: 0.34 ± 0.03; PM: 0.30 ± 0.03), SLA (GPM: 0.59 ± 0.07, PM: 0.61 ± 0.06), HCW (GPM: 0.58 ± 0.06, PM: 0.52 ± 0.07), REA (GPM: 0.48 ± 0.04, PM: 0.45 ± 0.05), FAT (GPM: 0.41 ± 0.05, PM: 0.30 ± 0.05), and MAR (GPM: 0.56 ± 0.07, PM: 0.51 ± 0.08). Additive genetic correlations between pairs of ultrasound and carcass traits were all between −0.31 and 0.81. The highest positive additive genetic correlations were between UW and UREA, UW and HCW, UW and REA, UREA and HCW, UREA and REA, UFAT and FAT, and between HCW and REA. The largest negative additive genetic correlations were between UREA and UPIMF, UFAT and SLA, UFAT and HCW, UPIMF and REA, and between REA and MAR. High similarity existed among predicted EBV and accuracies from GPM, GPMR1, and GPMR2 as well as high-rank correlations for sires, dams, and progenies. This indicated that the two reduced genotype sets were appropriate alternatives to the complete GPF250k set for genomic-polygenic evaluation and selection in this multibreed Angus-Brahman population. High EBV variability existed among animals of all Angus and Brahman percentages and no specific breed composition was overwhelmingly better or worse for any of the nine traits. This indicated that optimization of genetic progress through selection in multibreed Angus-Brahman populations should be based solely on genetic merit regardless of breed composition.