Genetic analysis for visual scores of bovines with the linear and threshold bayesian models

The objective of this work was to compare the estimates of genetic parameters obtained in single-trait and two-trait bayesian analyses, under linear and threshold animal models, considering categorical morphological traits of bovines of the Nelore breed. Data Of Musculature, physical Structure and conformation were obtained between years 2000 and 2005, from 3,864 bovines of the Nelore breed from 13 participant farms of the Nelore Brazil Program. Single-trait and two-trait bayesian analyses were performed tinder linear and threshold animal models. In general, the linear and threshold models were efficient in estimating genetic parameters for visual scores Under single-trait bayesian analyses. In the two-trait analyses, it was observed that: using continuous and categorical data, the threshold model provided greater estimates of genetic correlation than those of the linear model; with categorical data, the heritability estimates were similar. One major advantage of the linear models was its smaller requirements in the analyses processing time. In the genetic evaluation of animals for visual scores, the use of the linear or threshold model did not influence the classification of the animals, based on their predicted breeding values, which suggests that both models can be used in genetic improvement programs.