RESUMO
The use of mixed nonlinear models reduces the residual variance by including random effects, thereby improving the estimates in growth curves. Therefore, we tested distinct mixed nonlinear models (Richards, Brody, Gompertz, Logistic, and Von Bertalanffy) to model the growth curves of Zebu cattle herds raised on pasture in Brazil. In addition, we evaluated the diversity among breeds using hierarchical cluster multivariate analysis (full linkage) based on asymptotic weight (A), maturity rate (K), absolute growth rate (AGR), and absolute maturity rate (AMR) adjusted to 205, 365, 550, and 730 days of age. Body weight data from 311,834 males and females of Guzerat, Nellore, Tabapuã, and Polled Nellore breeds provided by the Brazilian Association of Zebu Breeders were used. The Brody model considering breed-sex fixed and animal random effects was the most suitable to describe the growth in the four breeds. The Guzerat and Tabapuã males showed increased precocity (higher K values) when compared to polled Nellore and Nellore males, which were characterized by higher A and AMR values. Females were more precocious than males (high K and low A values), particularly within Tabapuã and Guzerat breeds, respectively. Regardless of sex, the highest AGR values were observed from weaning up to 2 years of age in Tabapuã and Guzerat breeds. The cluster analyses revealed two groups: (1) Polled Nellore and Nellore males with high A, AGR_205, AGR_365, AGR_550, and AGR_730 values, represented by large adult animals, and (2) males and females of Guzerat and Tabapuã, characterized as animals of increased precocity. A subgroup encompassing males of Guzerat and Tabapuã was also observed, with intermediate values ââfor all parameters. This information is useful to help breeders and producers in strategies for feeding management and defining the most precocious beef cattle breed to increase the economic gains in pasture-based raising systems.
Assuntos
Dinâmica não Linear , Feminino , Masculino , Bovinos , Animais , BrasilRESUMO
We evaluate genotype × environment (G × E) interactions for age at first calving (AFC) and calving interval (CI) of Nellore cattle in northeastern Brazil using four hierarchical reaction norm models (HRNMs). The best-fit model for the traits was the one step heteroscedastic hierarchical reaction norm model. Heritability was close to zero in the worst environments and increased as the environments improved (from 0.06 to 0.12 for AFC and from 0.01 to 0.03 for CI). The correlations between the intercept and the slope of the reaction norms for CI and AFC were from medium to high magnitude (0.75 ± 0.10 and 0.90 ± 0.04, respectively), indicating that animals with higher average breeding values had the greatest responses to the improvement of environmental conditions. The variation in heritability indicates different response to selection according to the environment in which the animals of the population are evaluated. The G × E was evident in bulls with more female offspring. In conclusion, our data demonstrate that selection for AFC in medium- and high-level environments leads to higher genetic gains.