Comparison of traditional selection indices and the Multi-Trait Genotype Ideotype Distance Index (MGIDI) for improving rice (Oryza sativa L.) cultivars

Rice is a vital cereal crop that is crucial to ensuring global food security. Developing high-yield varieties is essential to feed the growing world population. Selecting the best-performing genotypes based on genetic gain is critical for improving rice production. In this process, various selection indices help identify superior genotypes. Thus, the current study aimed to select superior genotypes through a multi-trait selection index. Among the multivariate indices, the widely used classical Smith-Hazel selection index (SI) and the multi-trait genotype ideotype index (MGIDI) were used to evaluate the studied genotypes for eleven quantitative traits simultaneously. Three scenarios were formulated namely, retaining multicollinearity (SI-1), removing multi-collinearity (SI-2) and using a path analysis-based index (SI-3) for estimating the Smith-Hazel selection index. Using these indices, maximum selection gains of 9.2% with the Smith-Hazel index and 18.5% with the MGIDI index for grain yield were predicted. Notably, genotypes K5 and H7 were selected by the Smith (SI-1, SI-2, SI-3) and MGIDI selection index at a selection intensity of 15%. This shows that these genotypes exhibited strong performance across various traits and were classified as elite, highlighting their potential to contribute significantly to future breeding programs aimed at improving grain yield along with multiple traits in rice. The selection of these genotypes makes a valuable resource for developing productive rice varieties.