Multivariate analysis in rice (Oryza sativa L.) germplasms for yield attributing traits

Abstract

A study was conducted to evaluate the genetic diversity and relationships among sixty rice genotypes by assessing eleven morphological yield traits using principal component analysis (PCA) and cluster analysis at ANDUAT, Ayodhya (Uttar Pradesh), India. The results found significant variation among the genotypes, with some exhibiting higher values for certain traits which confirm genetic diversity. Cluster analysis revealed that Cluster V had the highest number of genotypes, while Cluster IV had the highest intra-cluster distance, suggesting that these genotypes would be useful for rice improvement. Principal component analysis revealed that the first two principal components, along with three other components, accounted for 75.11 percent of the total variability. Days to 50% flowering (DFF) in days was identified as the most accurate predictor of variability, followed by days to maturity (DM) in days, 1000 seed weight (TSW) in gm, and panicle length (PL) in cm. The principal component to be first (PC1) was linked with plant height (PH) and harvest index (HI) in gm, the second principal component (PC2) was linked with DFF and DM, the third (PC3) was linked with TSW and grains/panicle (GP) in number, the fourth (PC4) with panicles bearing per plant (PBP) in number and biological yield per plant (BY) in gramme, and the fifth principal component (PC5) is linked with PL and BY. The study identified several promising genotypes for various traits, including G.35, G.17, G.30, G.45, and G.46 for short plant height and G.60, G.40, G.54, G.55, and G.41 for high yield.