Análise de componentes principais para características de crescimento em bovinos de corte

Abstract

The objetive of this research was to study the relation among body weight and average daily gain in different ages, using principal components analysis. Data on 1663 birth weight (BW), weaning weight adjusted to 230 days (WW), yearling weight adjusted to 365 days (YW), long yearling weight adjusted to 550 days (LYW), average daily gain from birth to weaning (AGW), average daily gain from weaning to 365 days (AGY) and average daily gain from 365 days weight to 550 day weight (AGL) from crossbred animals, and data on 320 observations of the same traits from straightbreed Nellore animals were analysed. The model included the fixed effects of breed (only crossbred data), contemporary group, and linear and quadratic effects of age at calving. For body weight in different ages, the first principal component contrasted heavier and light animals after birth and explained about 79,0% and 78,0% of the variation for data on crossbred and Nellore animals, respectively. The second principal component compared heavier animals at weaning and yearling weight those at long yearling weight. It explained around 13,5% and 15,5% of the total variation, respectively, for data on F1 and Nellore breed. The major source of variation among animals on the two data set for body weight was due to differences in weight followed by differences in the ages they got those weight. For the traits expressed as average daily gain, the variation among animals was due to differences in birth season, the first principal component explaining about 52,0% of the variation on crossbred animals. This component compared animal with higher AGY with those with higher AGW and AGL. For data on Nellore breed, the first component explain about 56,0% of the total variation and also compared animals with higher AGY with those with higher AGW and AGL.