Análise de componentes principais dos valores genéticos dos parâmetros da curva de lactação de ovinos leiteiros

Abstract

The objective of this study is to use the principal components analysis to evaluate the variability and the relationships between the genetic values of a heptacarater analysis for parameters of the lactation curve of Lacaune sheep. Will be data generated by the genetic evaluation of the lactation curves of a dairy herd, composed of 3558 Lacaune animals, collected between 2008 and 2015, with lactation curves adjusted by the Wood model. Principal component analysis will performed to condense the information contained in breeding values of curve parameters in fewer orthogonal latent variables, called principal components, with minimal loss of information. By means of principal component analysis it was possible to find three main components that were responsible for 94.59% of the original data variation, as well as a selection index based on the main components and the genetic value of each animal, which could be used as a selection strategy in breeding programs, and because the parameters of the lactation curve of the herd have low to moderate heritabilities, through selection through index can increase the rates of gain, however if the objective is only improvements in the total accumulated production of the stock, the selection can be made through the third main component.