Multivariate Dissection of Yield-associated Traits in Greengram (Vigna radiata L. Wilczek) Using Principal Component, Path and Cluster Analysis

Abstract

The experimental study was conducted using 83 greengram genotypes during the rabi season (November, 2023 to February, 2024) at the Regional Agricultural Research Station (RARS), Lam, Guntur, Andhra Pradesh, India to study the yield-related traits and employed an integrated analytical framework comprising path coefficient analysis, principal component analysis (PCA), and hierarchical cluster analysis with heatmap visualization. Path coefficient analysis revealed that pods plant^-1 and days to maturity exerted a direct effect on the yield trait, indicating that these traits played a crucial role in influencing seed yield. In total, seven principal components were extracted, of which the first three PCs exhibited greater than 1 and collectively accounted for approximately 82.37% of total variability, suggesting that a substantial proportion of genetic diversity was captured within these components. The 83 genotypes were classified into 4 groups through hierarchical clustering. Group III comprised the largest number of genotypes (35), followed by Group I (25 genotypes) and Group II (21 genotypes), while Group IV contained only 2 genotypes. Additionally, a two-way clustered heatmap further elaborated on genotype-trait interactions and facilitated the identification of high-performing subsets for specific attributes. Genotypes belonging to group III demonstrated superior performance for pods plant^-1 and seed yield plant^-1 (g), while group I genotypes excelled in early maturity traits, making both groups promising candidates for targeted selection in future greengram improvement programmes.