Principal components for morphometric traits in Campolina horses

Abstract

Principal component analysis (PCA) was applied to evaluate the genetic variability and relationship between 15 morphometric traits in 91,483 Campolina horses, as well as to propose an index based on an aggregate genotype that promotes a particular selection objective. PCA was applied to the genetic (co)variance matrix among variables. After calculation of the principal components, the breeding values were estimated to obtain an index related to the component that explained most of the variation. The first principal component (PC1) accounted for 97.8% of the total additive genetic variance of the traits. PC1 contrasted animals in terms of body size (wither, back and croup heights, body length, and thoracic girth). PC1 traits showed higher heritabilities and positive and high genetic correlations. An index was obtained (HPC1) with the combination of the breeding values of different traits from PC1 which permitted the use of this index as an aggregate genotype to identify the best animals for selection. The second principal component (PC2) was much smaller and grouped traits related to head and neck morphometry, among others. These traits are commonly used for breed qualification, a fact explaining the small variation in this component. An evaluation of the effect of HPC1 on withers height in two-trait analysis was also made which provided positive genetic correlations of moderate to high magnitude (0.73–0.86), indicating that selection for this trait (important in Campolina horses) is accounted for in the index. The use of HCP1 could be considered as an important alternative to selection since it does not consider a single trait but rather a set of variables that capture body proportions.