Multivariate principal component analysis to evaluate growth performances in Malabari goats of India

• Published in: Tropical animal health and production 2020-09, Vol.52 (5), p.2451-2460
• Main Authors: Valsalan, Jamuna, Sadan, Tina, Venketachalapathy, Thirupathy
• Format: Article
• Language: English
• Subjects: Animals; Biomedical and Life Sciences; Body height; Body length; Body Weight; Genetic analysis; Genetic factors; Goats; Goats - genetics; Goats - growth & development; India; Life Sciences; Multiple regression models; Multivariate Analysis; Principal Component Analysis; Principal components analysis; Regression analysis; Regular Articles; Variability; Veterinary Medicine/Veterinary Science; Zoology
• ISSN: 0049-4747; 1573-7438
• DOI: 10.1007/s11250-020-02268-9

Evaluation of growth performances in Malabari goats was done with body weight and major morphometric traits, viz. body height, body length and chest girth at 6, 9 and 12 months, respectively. Data pertaining on 1082 Malabari goats spread over a period of 5 years (from 2014 to 2018) were used in the study. Least squares analysis of traits was done to adjust the effect of major significant non-genetic factors. Traits were analysed by using Varimax rotated principal component analysis (PCA) with Kaiser normalization to explain growth performances. Out of twelve principal components, PCA revealed four components explained about 67.78% of total variation. The first component (PC1) explained 28.02% of total variation. It was represented by significantly positive high loading of BH9, BH12 and BH6. The second component explained 15.090% of total variance with high loading of distance between BL9, BL6 and BL12. The third component explained 12.643% of variance and showed high component loadings for CG9, CG6 and CG12. The fourth factor accounted for 12.020% of total variability with comparatively higher loading WT12, WT9 and WT6. The communality ranged from 0.562 for BL12 to 0.848 for BH9. The body weight of adult Malabari goats was predicted using stepwise multiple regression of different interdependent morphometric traits and principal components. The multiple regression model with PC1 and PC2 was most precise with coefficient of determination ( R 2 ) value 74%. Therefore, the study revealed that extracted components revealed maximum variability of growth performances in Malabari goats which could be effectively used for selection and breeding programmes.