Genetic Dissection and Multi-Trait Selection in French Bean (Phaseolus vulgaris L.) Genotypes Using MGIDI and Multivariate Approaches

Productivity enhancement of French beans faces numerous challenges owing to the complexity of interactions between yield traits and the limitations of conventional univariate selection methods. The current selection index, like the Smith-Hazel index, suffers from multicollinearity and subjective weighting of market traits. Therefore, there is a need for a robust index capable of simultaneously selecting for multiple desirable traits to identify promising genotypes. To address this, a current field investigation was conducted during Rabi 2024–25 using 22 French bean primitive cultivars in a randomized block design. Eleven agronomic traits were assessed, and the Multi-Trait Genotype Ideotype Distance Index (MGIDI), a metric quantifying closeness of genotypes to an ideotype trait combination, was used to identify candidate lines. Significant genotypic variation (p < 0.01) was observed for all traits except pods per cluster. The seed yield per plot (1.3 m²) ranges from 71.0 g to 217.5 g, and the top-performing genotypes were Deep red kholar, Luli, Kholar light brown, and Kholar medium brown. High GCV, PCV, heritability, and genetic advance were observed in relation to plant height, seed index, and seed yield, indicating strong additive genetic effects. Seed yield per plot was positively and significantly associated with all traits except days to 50% flowering. Path analysis showed that seed index, pod length, and secondary branches have a high positive direct effect on seed yield. PCA explained 88.54% of total variation, with PC1 (51.18%) dominated by key yield traits, clearly separating high-performing genotypes (Deep red kholar, Luli). MGIDI identified Deep red kholar, TSG1, and Kholar Yellow large as closest to the ideotype with >80% selection success and maximum predicted gain for seed yield (32.35%). Overall, MGIDI proved superior in identifying balanced genotypes by overcoming multicollinearity and integrating multi-trait information. These findings provide a robust framework for ideotype-based breeding and accelerate the development of high-yielding, stable French bean cultivars.