Principal Component Analysis (PCA) and hierarchial clustering in tobacco (Nicotiana tabacum L.) for yield and yield attributing traits

Abstract

Multivariate statistical analysis techniques like Principal Component Analysis (PCA) and heirarchial clustering were used to evaluate Genetic diversity among 246 genotypes of Tobacco for six major yield and yield-related traits. The hierarchial clustering indicated that all the genotypes were clustered into eight major groups. The cluster III had the maximum number of genotypes with highest intra cluster distance and cluster IV and VIII showed maximum inter cluster distance indicating that the characterized tobacco genotypes in these clusters has high potential for various breeding goals. Principal component analysis and genotype by trait biplot analysis showed that the first four components accounted for 94.75 per cent of the total variation, with principal component 1 (PC1) accounting for 55.96 per cent and PC2 for 20.97 per cent of the total variation. The high yielding genotypes with other yield attributes identified in this study would offer valuable genetic material for breeding elite tobacco varieties.

Keywords

• tobacco genotypes
• pca
• rotated component matrix
• eigen value