PROMOTING ACCESS TO AFRICAN RESEARCH

South African Journal of Animal Science

Log in or Register to get access to full text downloads.

Remember me or Register



Application of random regression models to the genetic evaluation of cow weight in Bonsmara cattle of South Africa

KA Nephawe

Abstract


Repeated records (n = 58, 295) of Bonsmara cows were used to evaluate mature cow weight (CW), using random regression animal models (RRM). Cows (n = 11, 847), age two to 11 years (AY), were weighed at weaning of their calves. Data were analysed with the Restricted Maximum Likelihood (REML) procedure, using orthogonal (Legendre) polynomials on age in months (AM). The model included fixed regression on AM (range from 30 to 138 mo) and the effect of herd-measurement date concatenation. Random parts of the model were RRM coefficients for additive and permanent environmental effects, while residual effects were modelled to account for heterogeneity of variance by AY. Estimates of variances increased with cow age, though tended to be flat at older ages. Heritability estimates ranged from 0.39 to 0.47, and were comparable with estimate of 0.41 obtained using a simple repeatability model (SRM). Estimates of genetic correlations were greater than 0.82 among measures of weight at all ages. The resulting covariance functions were used to estimate breeding values of each animal along the age trajectory. Genetic trends for CW over the years showed only a slightly increasing pattern, suggesting that CW did not change much, and was similar whether SRM or RRM was used. Results suggest that selection for CW could be effective and that RRM could be useful for the National Genetic Evaluation of CW in Bonsmara cattle. However, given the complexity of the RRM, for practical purposes a SRM might be an acceptable approximation for prediction of breeding values.
Key Words: Covariance functions, Mature weight, Genetic parameters, Beef cattle
South African Journal of Animal Science Vol.34(3) 2004: 166-173



http://dx.doi.org/10.4314/sajas.v34i3.3960
AJOL African Journals Online