Identification of anappropriatelinear model to evaluate pre-weaning traits in colombian creole cattle Blanco OrejinegroL

Identification of an appropriate linear model to evaluate pre-weaning traits in colombian creole cattle Blanco Orejinegro

Abstract

The Blanco Orejinegro (BON) Creole breed has an adaptation process of more than 500 years to the environmental conditions of Colombia, and it is characterized by being a dual-purpose breed used for both milk and meat production, so it is a very important biological heritage that must be studied. The goal of this study was to identify an appropriate linear model to evaluate pre-weaning traits in BON cattle. Weighing information was collected and filtered from four BON cattle herds. The evaluated traits were weight at 4 months (W4M), weight at weaning (WW) and daily weight gain between 4 months and weaning (DWG4M-W). Nine linear models were evaluated in which were included the following factors as fixed effects: sex, herd, month of weighing or month of birth, calving number, time of weighing or birth (dry and rainy weather), age (covariate, fixed effect and adjusted by regression), year of weaning or year of birth and, contemporary group (CG) formed by sex and herd for DWG4M-W and sex, herd and year of weighing for W4M and WW, with a minimum of five observations in each CG. To identify the most appropriate linear model for each trait, the value of AIC (Akaike information criterion, BIC (Bayesian information criterion), determination coefficient (R²) and sum of squares of error (SCE) were used. The most suitable model for all the traits was the one that involved the contemporary group and age as a fixed effect for W4M and age as a covariate for WW.