The objectives of this study were to estimate heritabilities and genetic correlations for milk (MY), fat (FY) and protein (PY) yields and for percentages of fat (%F) and protein (%P) in milk of Holstein cows in Paraná State. Data consisted of 117,082 lactation records for MY, FY and PY and 116,747 records for %F and %P from herds supervised by the Milk Recording Service of the Holstein Breeders Association of Paraná from January 1992 to December 2003. Covariance components and genetic parameters were estimated by Bayesian inference using an animal model including the fixed effects of contemporary group, number of milking, linear and quadratic effects for calving age and the additive genetic, permanent environmental and residual random effects. Averages and respective standard deviations for MY, FY and PY (kg) were respectively 8,181.23 ± 1,903.51, 270.88 ± 67.58 and 249.01 ± 55.65 and for %F and %P they were 3.33 ± 0.45 and 3.06 0.19, respectively. Heritability estimates for MY, FY, PY, %G and %P obtained from single-trait analyses were 0.26, 0.28, 0.25, 0.60 and 0.58, respectively. They indicate that these traits may be improved by selection. Genetic correlation estimates between MY and FY, MY and PY, FY and PY, MY and %F, MY and %P and for %F and %P were 0.56, 0.89, 0.67, -0.42, -0.39 and 0.56, respectively. Selection for increasing MY will result in positive correlated genetic gains for kg of fat and protein but decrease their percentages in milk.
Bayesian inference; genetic correlation; Gibbs sampler; heritability; milk components
