Comparison of single-trait and multi-trait animal models to estimate components of (Co) variance and genetic growth parameters in the bovine breed Costeño con Cuernos

Abstract

The study aimed to estimate the components of (co) variance and genetic parameters for growth traits as birth weight (BW), weaning weight (WW) and weight at 16 months of age (W16) of calves of the creole breed Costeño con Cuernos, born between 1989 and 2011 in a herd located in the northeast of Colombia. For this, 3060, 2594 and 1990 BW, WW and W16 records were analyzed, respectively. Single-trait (UNI) and multi-trait (MT) models were used. Six variants were evaluated with the UNI model, in all cases the random effect of the animal was included and the rest of the models alternated the maternal effect, the effect of the permanent environment and the relationship between the direct and maternal effect. Goodness of fit was determined using the information criteria of Akaike (AIC), Bayesiana (BIC) and LogL. For BW and WW models involving direct, maternal, and permanent maternal environment effects showed the best fit, and for W16 the best fit model was the one that only considered the direct effect. The heritability values of the UNI and MT models were similar for BW (0.14 and 0.16, respectively). For WW and W16, the MT model, in relation to the UNI, increased h²_a by 42% (0.27 vs. 0.19; 0.31 vs. 0.21, respectively), as well as h²_T for WW by 34% (0.31 vs. 0.23). The genetic variability found allows the application of a selection process and it is recommended to use the multi-trait models for the genetic evaluation of the herd.