The use of subjectively assessed muscular and skeletal traits on live cattle to aid in differentiation between animal genetically divergent in carcass kill out metrics

Abstract

Subjective linear scoring of live beef cattle is routinely undertaken as part of breed society regulations or as part of national breeding programs; linear scores describe biological extremes of animals for a range of different traits reflecting muscularity, skeletal development, and functionality. The objective of the present study was to quantify the usefulness of these linear scores measured on live growing beef cattle to predict genetic merit for kill out (KO) percent and the difference between live-weight at slaughter and carcass weight (herein known as KO difference). The data used consisted of linear scores for 16 traits on up to 67,167 cattle as well as KO information on 31,827 cattle; 1,166 animals had records for both sets of traits. Variance components were estimated using univariate animal linear mixed models while covariance components between the linear scores and the KO traits were estimated using a series of bivariate sire linear mixed models. In an additional series of analyses, the KO metrics were adjusted phenotypically for differences in live-weight at slaughter through its inclusion as a covariate in the statistical model. Heritability estimates of the linear scores varied from 0.06 (width at pins) to 0.37 (development of hind-quarter); the heritability of KO percent and KO difference were estimated to be 0.53 and 0.37, respectively. Both the phenotypic and genetic correlations between the muscular type traits and KO percent were moderately positive, albeit the genetic correlations were stronger. The phenotypic correlations ranged from 0.27 (development of inner thigh) to 0.37 (development of hind quarter) while the genetic correlations varied from 0.40 (development of inner thigh and development of loin) to 0.60 (development of hind quarter); in all cases, adjusting for differences in live-weight at slaughter had minimal impact on the estimated correlations. With the exception of depth of rump, the phenotypic and genetic correlations between the skeletal traits with KO percent were all close to zero (≤|0.24|) irrespective of whether or not differences in live-weight at slaughter were accounted for. While the genetic correlations between the muscular traits and KO difference not adjusted for differences in live-weight at slaughter were all close to zero (≤|0.30|), the correlations strengthened (≥|0.39|) once adjusted to a common live-weight at slaughter. The opposite was true for the genetic correlations between the skeletal traits and KO difference. In all, the results suggest that the muscular linear scores assessed subjectively on live animals at, on average, 10 months of age are a useful genetic (and phenotypic) predictor of KO percent at, on average, 21 months of age, but also the quantity of live-weight that does not end up as carcass, once adjusted to a common live-weight.