Productive and reproductive traits in Murrah breed from Brazil

Major objective of this study was to estimate heritability and genetic correlations between milk yield (MY) and calving interval (CI) and lactation length (LL) in Murrah buffaloes using Bayesian inference. The database used belongs to the genetic improvement program of four buffalo herds from Brazil. To obtain the estimates of variance and covariance, bivariate analyses were performed with the Gibbs sampler, using the program MTGSAM. The heritability coefficient estimates were 0.28, 0.03 and 0.15 for MY, CI and LL, respectively. The genetic correlations between MY and LL was moderate (0.48). However, the genetic correlation between MY and CI showed large HPD regions (highest posterior density interval). Milk yield was the only trait with clear potential for genetic improvement by direct mass selection. The genetic correlation between MY-LL indicates that indirect selection using milk yield is a potentially beneficial strategy. The interpretation of the estimated genetic correlation between MY-CI is difficult and could be spurious.