Multivariate analysis of local pig populations in the rural areas of High-Katanga province, DR Congo

Abstract

Morpho-biometric characteristics of 342 local pigs from the rural areas of Kasenga and Kambove in Haut-Katanga province were studied with the objective to analyzing genetic variability, determining population structure, establishing phylogenetic relationships between the pig populations studied and identifying probable genetic types. The data collected were subjected to principal component analysis (PCA), hierarchical clustering (HCA) and discriminant factor analysis (DFA). The main results showed that the first two principal components (F1 and F2) explained more than 61.5% of the total phenotypic variability observed in the pig population studied. The highly correlated variables are height at withers, shoulder length, body length, head length, hip width, pelvis length, neck circumference, muzzle length, tail length, hind leg length, hock length, foreleg length. The F1 axis is clearly related to height at withers, body length, shoulder length, and neck circumference, while F2 is mainly related to weight and foreleg length. Height at withers, shoulder length and body length seem to be the most discriminating variables. The study population would consist of three genetic types, which can be grouped into two subgroups based on intra/inter population variation and genetic distances. The quantitative parameters studied, and sex are the variables that best characterize the pig populations (p<0.05). The local pig is therefore an animal genetic resource with high breeding potential due to its biodiversity.