Identifying genetically diverse and high-yielding wheat genotypes is crucial for increasing productivity and addressing food security concerns in Bangladesh. The objective of this study was to assess the genetic diversity and agricultural potential of 49 wheat genotypes in Bangladesh using multivariate analyses. Genotypes were analyzed for 31 morpho-physiological, phenological, and yield-related parameters. Cluster analysis divided the genotypes into seven distinct clusters, which exhibited significant intercluster variation. Clusters V and VI are the most promising regarding higher grain yield, earlier physiological maturity, and efficient biomass accumulation. On the other hand, productivity rates were low in Clusters II and IV. Principal component analysis (PCA) explained 86.46% of the total variation through nine principal components, of which PC1 accounted for 24.29% of the variability, which was mainly related to grain yield and absorbed photosynthetically active radiation (PAR) and SPAD values. Discriminant Function Analysis (DFA) identified grain fresh weight and grain yield as the most discriminative traits, effectively separating high- and low-yielding genotypes. Mahalanobis D² analysis confirmed significant genetic divergence between clusters, especially between clusters IV and VII, indicating potential for hybridization. The classification accuracy from DFA was 96.4%, which verifies the robustness of the clustering results. Representative genotypes were identified for each cluster to guide future breeding strategies. Overall, this study shows that multivariate methods are effective tools for characterizing wheat germplasm and identifying elite genotypes for breeding development. These results provide a strong foundation for future breeding programs aimed at contributing to national food security in Bangladesh under changing climate conditions.
