Genetic variability, heritability, and trait associations for yield and grain dimensions in diverse rice genotypes

Abstract

Rice ( Oryza sativa L.) is a critical staple crop, necessitating continuous genetic improvement to enhance yield and quality traits to meet global food demands. This study aims to identify high‐yielding genotypes with desirable quality traits and elucidate the relationships among traits to inform breeding strategies. Twenty‐seven rice genotypes were evaluated for genetic variability, heritability, genetic advance, and trait associations. Nine agronomic traits were measured: grain yield, days to maturity, plant height, productive tiller number, 1000‐grain weight, grain length, grain width, grain shape, and awn length during the 2021–2023 cropping seasons. Analysis of variance revealed highly significant differences ( p  < 0.01) among genotypes for all traits, demonstrating pronounced phenotypic variations that suggest a diverse genetic background among the studied genotypes. Grain yield ranged from 4.38 t ha ^{− 1} (Gerdeh) to 7.73 t ha ^{− 1} (Shiroudi), with a mean of 6.052 t ha ^{− 1} . Broad‐sense heritability ( H ² ) was high (0.926–1.0) for all traits. Traits with both high heritability and high genetic advance as percentage of mean (GA% > 20%)—namely, grain shape, productive tiller number, grain yield, grain width, and grain length—are likely to respond effectively to direct phenotypic selection. Correlation analysis showed significant positive associations between grain yield and days to maturity ( r  = 0.532, p  < 0.01), 1000‐grain weight ( r  = 0.266, p  < 0.05), grain length ( r  = 0.269, p  < 0.05), and grain shape ( r  = 0.269, p  < 0.05), but a negative correlation with awn length ( r  = −0.406, p  < 0.01). Path analysis identified days to maturity (direct effect = 0.341, p  < 0.001) and grain shape (0.241) as key yield contributors. Hierarchical clustering grouped genotypes into four clusters, with Cluster 4 (including Shiroudi, Keshvari, and Neda) containing high‐yielding genotypes. These results provide clear guidance for designing efficient breeding programs through targeted selection of superior genotypes and trait‐based strategies. Shiroudi and Cluster 4 genotypes are promising candidates for breeding high‐yielding, quality rice varieties.