Multi-trait selection indices for identifying elite rice genotypes in rice breeding programs

Rice (Oryza sativa L.) serves as a staple food for nearly half of the global population, with rising demand necessitating significant advancements in productivity. Traditional selection methods that focus solely on yield often fail to account for the complex interplay of agronomic and grain quality traits. The integration of multiple selection indices in breeding enhances efficiency by simultaneously evaluating important traits, aiding in informed decision-making, balancing desirable traits, and accelerating the development of high-performing varieties. This study aimed to evaluate the efficiency of various multi-trait selection indices, namely the Multi-Trait GenotypeIdeotype Distance Index (MGIDI), Genotype by Yield and Trait biplot (GYT), Linear Phenotypic Selection Index (LPSI), and the Elston Index, in identifying elite rice genotypes for breeding programs. A total of 110 genetically diverse rice germplasm lines were evaluated using a randomized block design during the Rabi season of 2023–24. Key agronomic and grain quality traits were assessed, with statistical analyses, including ANOVA and correlation studies, conducted to interpret the result. Among the indices, MGIDI demonstrated the highest selection gains (16.9%) for yield, while other indices demonstrated variable efficiencies across different traits. Traits such as the number of grains per panicle and productive tillers exhibited positively correlations with yield, whereas negative selection for plant height and days to maturity posed challenges. Notably, genotypes BMDK-2-2-8-2, JR 13, A 67, and CR 4376-1-1-1-2-2-1 were consistently selected across indices, reflecting their superior trait performance across multiple traits. Combining several indices improves the breeding process by enabling the selection of genotypes with traits such as nutrient-use efficiency and drought tolerance, thereby improving rice yield under challenging conditions, such as lowfertility soils or drought stress. These findings highlight the importance of multi-trait indices in optimizing genetic gains and improving breeding efficiency. Notably, MGIDI emerged as the most effective tool, providing a comprehensive approach to integrating traits, making it indispensable for rice breeding programs.