Development of Angus SteerSELECT: a genomic-based tool to identify performance differences of Australian Angus steers during feedlot finishing: Phase 1 validation

Abstract

Context Genomic-based technologies are allowing commercial beef producers to predict the genetic merit of individual animals of unknown pedigree with increased ease and accuracy. Genomic selection tools that can accurately predict the feedlot and carcass performance of steers have the potential to improve profitability for the beef supply chain. Aims To validate the ability of the Angus SteerSELECT genomic product to predict differences in performance of Australian Angus steers, in terms of carcass weight, marbling score, ossification score and carcass value, using a short-fed (100 days) or long-fed (270 days) finishing protocol at a commercial feedlot. Methods A reference population of 2763 Australian Angus steers was used to generate genomic prediction equations for three carcass traits, namely, carcass weight, marbling score and ossification. The accuracy and bias of genomic predictions of breeding values were then evaluated using a validation population of 522 Angus steers, either short- or long-fed at a commercial feedlot, by comparing breeding values to measured phenotypes. The potential economic benefits for feedlot operators when using Angus SteerSELECT were estimated on the basis of the ability of the tool to predict the carcass value of steers in the validation population. Key results The accuracy of genomic predictions of breeding values for carcass weight, marbling score and ossification score were 0.752, 0.723 and 0.734 respectively. When steers were ranked in quartiles for predicted carcass value, calculated using genomic predictions of breeding values for carcass weight and marbling score, the least-square mean carcass value for steers in each quartile, from bottom 25% predicted performers to top 25% predicted performers, were estimated at A$1794, A$1977, A$2021 and A$2148 for short-fed steers and A$3546, A$3780, A$3864 and A$4258 for long-fed steers. Differences in the carcass value least-squares mean between the bottom and top quartile were highly significant (P < 0.001) for both short-fed and long-fed steers. Conclusions Genomic prediction equations used in Angus SteerSELECT can predict differences in carcass weight, marbling score, ossification score and carcass value in both short-fed and long-fed Australian Angus steers. Implications Genomic selection tools that can predict differences in performance, in terms of growth and carcass characteristics, of commercial feedlot cattle have the potential to significantly increase profitability for the beef supply chain by improving the quality and consistency of the beef products they produce.