Skip to main content
ARS Home » Plains Area » Clay Center, Nebraska » U.S. Meat Animal Research Center » Genetics and Animal Breeding » Research » Publications at this Location » Publication #276565
Title: The accuracies of DNA-based estimates of genetic merit derived from Angus or multibreed beef cattle training populations

Author
item WEBER, KRISTINA - University Of California
item DRAKE, DANIEL - University Of California
item TAYLOR, JERRY - University Of Missouri
item GARRICK, DORIAN - Iowa State University
item Kuehn, Larry
item Thallman, Richard - Mark
item SCHNABEL, ROBERT - University Of Missouri
item Snelling, Warren
item Pollak, Emil
item VAN EENENNAAM, ALISON - University Of California

Submitted to: Journal of Animal Science
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 7/24/2012
Publication Date: 12/1/2012
Citation: Weber, K.L., Drake, D.J., Taylor, J.F., Garrick, D.J., Kuehn, L.A., Thallman, R.M., Schnabel, R.D., Snelling, W.M., Pollak, E.J., Van Eenennaam, A.L. 2012. The accuracies of DNA-based estimates of genetic merit derived from Angus or multibreed beef cattle training populations. Journal of Animal Science. 90(12):4191-4202.

Interpretive Summary: With the advent of large scale genotyping arrays, national beef cattle genetic evaluation incorporating information from markers across the entire bovine genome is now possible. However, due to different resource populations used to estimate marker effects, there are now multiple prediction models available to develop molecular breeding values (MBV; genomic estimates of genetic merit) in cattle. The effectiveness of these different models likely varies according to the relationship of the target population to the populations used to predict marker effects; the resulting usefulness of each prediction is therefore often unknown. Our objective was to test the effectiveness of several different marker prediction equations used to estimate MBV in a set of commercial Angus bulls. Prediction equations for weaning weight, hot carcass weight, marbling score, rib-eye muscle area, and average daily gain were evaluated by estimating the genetic correlation between the resulting MBV and the phenotypes for each trait. This genetic correlation is a direct measure of prediction accuracy. Results suggested that marker predictions developed using purebred Angus populations were more predictive of progeny performance in this commercial Angus population than those trained on multiple breed and crossbred populations.

Technical Abstract: Several organizations have developed prediction models for molecular breeding values (MBV) for quantitative growth and carcass traits in beef cattle using BovineSNP50 genotypes and phenotypic or EBV data. MBV for Angus cattle have been developed by IGENITY, Pfizer Animal Genetics, and a collaboration between researchers from Iowa State University and the University of Missouri-Columbia (ISU/UMC). The U.S. Meat Animal Research Center (USMARC; Clay Center, NE) has also developed MBV for 16 cattle breeds using 2 multibreed populations, the GermPlasm Evaluation program (GPE) and the 2,000 Bull Project (2K*ALL), and 2 single breed subpopulations of the 2,000 Bull Project, Angus (2K*AN) and Hereford (2K*HH). In this study, these MBV were assessed relative to commercial ranch EBV estimated from the progeny phenotypes of Angus bulls naturally mated in multisire breeding pastures to commercial cows: 121 for USMARC MBV, 99 for ISU/UMC MBV, and 29 for IGENITY and Pfizer MBV (selected based on number of progeny carcass records). Five traits were analyzed: weaning weight (WW), hot carcass weight (HCW), marbling score (MS), rib-eye muscle area (RE), and, for IGENITY and Pfizer only, feedlot average daily gain (ADG). The average accuracies of MBV across traits were: IGENITY 0.38 ± 0.05, Pfizer 0.61 ± 0.12, ISU/UMC 0.46 ± 0.12, GPE 0.16 ± 0.04, 2K*ALL 0.26 ± 0.05, 2K*AN 0.24 ± 0.04, and 2K*HH 0.02 ± 0.12. Angus-based MBV (IGENITY, Pfizer, ISU/UMC, and 2K*AN) explained larger proportions of genetic variance in this population than GPE, 2K*ALL, or 2K*HH MBV for the same traits. In this data set, IGENITY, Pfizer, and ISU/UMC MBV were predictive of realized performance of progeny, and incorporation of that information into national genetic evaluations would be expected to improve EPD accuracy, particularly for young animals.