Growth curves and genetic parameters in Nelore animals estimated by random regression models

Abstract

The objective of this work was to estimate growth curves and genetic parameters from birth to 650 days of age of Nelore cattle raised in pasture in two production regions of the Mato Grosso do Sul State, Brazil (233,835 weight records from 47,459 cattle were analyzed). Genetic parameters were determined by random regression using Legendre orthogonal polynomials of cubic order, and age at weighing was considered in the model as a fixed effect to model the average growth trajectory. In the models, the effects of the contemporary group were considered as fixed and, as covariates, the animal age at weighing and the cow age at calving were nested in the animal age class (linear and quadratic effects), forming eight age classes. All models included the direct genetic additive, maternal genetic, and animal permanent environment as random effects, and the most appropriate model to describe the studied effects was defined according to the AIC and BIC criteria. Heritability estimates for birth weight varied between the two production regions, Campo Grande-Dourados (R1) and Alto Taquari-Bolsão (R2) and R1 (0.36 ± 0.02) and R2 (0.28 ± 0.03), and there were variations in the estimates at advanced ages. In both regions, the highest heritability values at 650 days of age were 0.47 ± 0.03 and 0.65 ± 0.02 for R1 and R2, respectively, with high heritability reflecting the high values of additive genetic variance. The random regression methodology was efficient in estimating growth curves and genetic parameters. Growth curves were different when they were estimated separately by sex, birth season, and production region. Genetic parameters estimated separately by region indicate differences in additive genetic variance, maternal additive, and animal permanent environment for weights up to 650 days of age.