Principal component analysis of yield and associated traits in field Pea (Pisum sativum L.) across multiple environments

Abstract

The characterization of genetic variability across diverse environments constitutes a prerequisite for effective selection strategies in field pea (Pisum sativum L.) breeding programs. The present investigation employed principal component analysis (PCA) to evaluate 70 field pea genotypes across four distinct environments and, in a pooled analysis, to examine yield and associated agronomic traits. In the pooled dataset, three principal components with eigenvalues exceeding unity accounted for 76.57% of the total phenotypic variation, whereas three to four components explained 66.60–77.96% of the variation across individual environments, thereby demonstrating the multivariate architecture of yield expression. The first principal component (PC1) consistently accounted for the highest proportion of variability (37.28–42.06%) and was predominantly influenced by seed yield per plant, biological yield, pods per plant, effective nodes, total nodes, and plant height, thereby underscoring biomass accumulation and sink capacity as principal determinants of genetic divergence. The second principal component (PC2) was predominantly associated with phenological and partitioning traits i.e., days to 50% flowering, days to maturity, harvest index, pod length, and number of seeds per pod, whereas PC3 and PC4 captured variation attributable to seed size and reproductive attributes. Genotypes including JFP-27, Triple Branching, DDR-44, JFP 99-25, NDVP-4, and JM-6 demonstrated multi-trait superiority across principal components, indicating their potential utility in breeding programs. The findings corroborate the efficacy of PCA in elucidating complex trait interrelationships and identifying superior genotypes for multi-trait improvement in field pea breeding programs.