Alternative models of genetic evaluation of reproductive traits for local goat population under arid conditions

Abstract

The success of small ruminant breeding programs depends on appropriate models for genetic evaluation of breeding animals on traits showing a genetic association between different reproductive phenotypes. The objective of this study is to determine the environmental and genetic components that synthesize the reproductive performance of local goats in arid conditions, in order to make the foundations for future genetic improvement of this population. Two reproductif traits were studied : the interval between kidding and the litter size at birth. Three models were used for each case, univariate and bivariate analyzes, which differ in random effects included in the model, only the additive genetic effects, only the permanent environmental effects, or both effects simultaneously. The obtained results suggest that the Bi-variate model including only the permanent environmental effects of the goat is the most adequate to model the reproductive performance of local goat. Direct genetic variances are mainly higher using the uni-variate model than the Bivariate model. The environmental variance estimates obtained are the same for both models. The heritabilities recorded from the uni-variate model are slightly lower than the estimates from the Bi-variate model. It seems imperative to consider in animal model, all the genetic and environmental effects, that intervene on the determinism of reproductif traits.

Keywords: local goat, improvement, evaluation, uni-variate model, Bi-variate model, reproduction