COMPARISON OF MODELS FOR THE ESTIMATION OF VARIANCE COMPONENTS FOR GROWTH TRAITS OF REGISTERED LIMOUSIN CATTLE

Abstract

Six models to estimate genetic parameters for birth weight (BW), weaning weight adjusted to 205 days (W205), and yearling weight adjusted to 365 days (W365) were compared. Model A included direct genetic effects. Model AP allowed for direct genetic and permanent environmental effect of the dam. Model AM included direct genetic and maternal genetic effects. Models AMC and AMP were the same as Model AM but they also allowed for the covariance between direct and maternal genetic effects, and the common environmental effect due to the dam, respectively; and Model AMCP was fitted for all three random effects plus the covariance between direct and maternal effects. Models were compared using the likelihood ratio text. The AMC model was selected to be the most appropriate for BW and W205, whereas Model A was chosen for W365. When maternal effects were included, direct genetic variance and direct heritability estimates were reduced for BW and W205. Direct heritability estimates with appropriate models were: 0.13, 0.21 and 0.20 for BW, W205 and W365. Heritability of maternal effects with appropriate models was: 0.15 and 0.32 for BW and W205, and direct-maternal genetic correlations with appropriate models were: -0.67 and -0.69 for BW and W205, respectively.