Data from Nellore beef cattle breed were used to study the influence of variance heterogeneity on genetic evaluation. Adjusted weights at 365, 450 and 550 days of age were classified as high and low classes of standard deviation, based on phenotypic standard deviation of herds in each weight. At multiple trait analyses, where the same weight was considered a distinct characteristic regarding each class of standard deviation, genetic and residual variances increased as the standard deviation of the class increased. Genetic correlations between the same weight, in high and low phenotypic standard deviation classes were 0.71, 0.80 and 0.84 for weights at 365, 450 and 550 days, respectively. Spearman correlations between breeding values obtained from multiple trait analyses and general analysis (without classes), and between classes of standard deviations, lowered as the intensity of selection on sires increased. The presence of heterogeneity of variances has greater impact on the genetic evaluation of breeding under high intensity of selection, it is interesting your consideration in the process of genetic evaluation.
beef cattle; genetic evaluation; genetic parameters; genotype × environment interaction; selection
