Random regression models to traits of growth in the Nellore breed in Mato Grosso state

In this study 138,976 records of live weight between 60 to 610 days of age, from 27,327 Nellore cattle breed, from herds in Mato Grosso State were used in order to describe the genetic variability and to estimate genetic parameters for the live weight at different ages, using random regression models. The model included the fixed effects of contemporary group and age of cow at calving as covariate, random effects of direct additive genetic, maternal genetic, animal and maternal permanent environmental and temporary environment effect. The most appropriate random regression model employed the covariance function with fourth order polynomials to describe the variability of all effects and two residual variance classes. Estimates of direct additive genetic variance and animal permanent environment increased with the age of the animals. Maternal genetic variances and maternal permanent environment exhibited similar behavior, with higher values in pre weaning. The estimated heritability coefficients ranged from 0.25 to 0.43, with higher values at older ages in the growth trajectory of the animals. These results showed the presence of sufficient genetic variability to obtain significant genetic gain through selection, especially after weaning. The results for the direct additive genetic correlation exhibited low correlations between weights in initial and final ages, however, highly correlated weights between nearest ages. Genetic correlation estimates between weaning with weights up to 610 days of age were high and positive and indicate that most of the genes responsible for higher weights in this period are the same.