Estimation of Variability in Body Morphometric Measurements Using Principal Component Analysis in Mehsana Goats
Jagdish D. Chaudhari *
Department of Animal Genetics and Breeding, College of Veterinary Science and Animal Husbandry, Kamdhenu University Erstwhile S.D. Agricultural University, Sardarkrushinagar, Gujarat -385 506, India.
Jay Prakash Gupta
Department of Animal Genetics and Breeding, Bihar Veterinary College, Patna – 800014, India.
Pravin B. Purohit
Department of Animal Genetics and Breeding, College of Veterinary Science and Animal Husbandry, Kamdhenu University Erstwhile S.D. Agricultural University, Sardarkrushinagar, Gujarat -385 506, India.
Amit K. Srivastava
Department of Livestock Production and Management, College of Veterinary Science and Animal Husbandry, Kamdhenu University Erstwhile S.D. Agricultural University, Sardarkrushinagar, Gujarat - 385 506, India.
*Author to whom correspondence should be addressed.
Abstract
Aims: The study was conducted on Mehsana breed of goats which were selected randomly from their breeding tract in Gujarat State, India with an aim to estimate the variability in body morphometric measurements using Principal Component analysis.
Study Design: The principal component analysis (PCA) was used to characterize the most important component of variables from morphometric traits.
Place and Duration of Study: Department of Animal Genetics and Breeding, College of Veterinary Science and Animal Husbandry, Kamdhenu University erstwhile S.D. Agricultural University, Sardarkrushinagar, Gujarat, India-385 5063 during 2019-20.
Methodology: A total of 118 Mehsana goats were randomly selected for this study from 13 villages of 3 districts. The data on body weight (BW) as well as measures of different body parts (in centimetre) namely, Heart girth (HG), Height at Withers (HW), Height at Rump (HR), Shoulder width (SW), Body length (BL), Tail length (TL) and Paunch girth (PG) were collected.
Results: The accuracy (R2) was estimated up to 70% (Heart girth) when body weight was predicted using single variable. Increase in coefficient of determination (R2) was observed as the number of independent variables in the equation increases. Inclusion of all the independent variables in prediction equation fetched accuracy up to76%. Out of eight principal components, the first two components explained cumulative percentage of variance of 70.967%. The first principal component (PC1) contributed 57.875% of total variation whereas; second component (PC2) explained 13.092% of the total variance. Varimax rotation was used for rotation of principal factors through the transformation of the factors to approximate a simple structure and inferred that HW (0.883), HR (0.848), BL (0.807) and HG (0.761) had significantly higher loadings in the first PC1 and the PC2 mainly contain the significantly higher loadings for TL (0.794).
Conclusion: The variance of about 70.967% was explained by the two major components, out of total eight principal components extracted which indicated that PCA can be used effectively for multivariate analysis in Mehsana goats based body measurers and body morphometric traits.
Keywords: Mehsana goats, morphometric traits, principal component, regression equation