Genetic parameters and trends for milk production of blond-faced Latxa sheep using Bayesian analysis.

The genetic progress attained with the breeding scheme of the blond-faced Latxa dairy sheep in the Spanish Basque Country was assessed by a Bayesian approach based on the marginal posterior distributions of parameters achieved via Gibbs sampling. The data file included 49,056 milk yield records of 22,363 ewes. Normal distributions were assumed for EBV and fixed effects, and scaled inverted chi-square distributions were assumed for variance components or additive, permanent environment, and residual effects. Under vague priors for variance components, the posterior means (SD) for heritability and permanent environment coefficient for milk yield at 120 d were 0.22(0.01) and 0.20(0.01). An important effect on the milk yield (41.6% of the mean) was evident for a genetic group of imported rams. Selection was effective for sheep that were bred by AI and that had known parents. The robustness of results for effect of genetic group and genetic trends was validated using a mildly informative prior constructed from variance components estimated with the related black-faced Latxa breed.