Genotype-environment interaction in the genetic variability analysis of reproductive traits in Nellore cattle

Abstract

The research aim was to evaluate the genotype-environment interactions (GxE) through a study of the reaction norms of Nellore cattle raised in the North and Northeast regions of Brazil at the age at first calving (AFC), with scrotal circumference at 365 and 450 days of age (SC365 and SC450), gestation length (GL) and accumulated productivity (ACP). Bayes hierarchical models with defined parameters at structured levels or stages were used. The Markov chain Monte Carlo method was used to obtain genetic parameter estimates. The environmental gradient of each trait was defined using a 99% credibility interval of the contemporary group's solutions. The animal model showed a better fit for AFC data, and the two-step reaction norms model was the best for the other characteristics. Heritabilities were 0.21 +/- 0.01 for AFC; and ranged from 0.27 +/- 0.04 to 0.58 +/- 0.02, 0.26 +/- 0.05 to 0.57 +/- 0.02, 81 +/- 0.03 to 0.03 +/- 0.02, 0.05 +/- 0.02 to 0.86 +/- 0.03 along the environmental gradient for SC365, SC450, GL and ACP, respectively. Correlations between intercept and slope of the reaction norms allied with reaction norms graphic plots of the ten sires with the greatest number of calves suggest the presence of GxE complex for SC365 and SC450, and thcE scale effect for GL and ACP. However, high correlations between the environments indicate that the GxE for SC450 and ACP is of a low magnitude and its inclusion in genetic evaluations would not influence the population selection.