Conventional breeding is vital in improving yield-related traits and developing high-yielding cultivars. In the presented study, an evaluation of 15 F₁ hybrids compared them to their six parental genotypes for genetic variation of various traits. The results showed that the analysis of variance revealed significant (P ≤ 0.01) differences among the cross combinations and parental cultivars and among the parents and crosses. The mean sum of squares of parents vs. crosses also indicated substantial (P ≤ 0.01) for all the characteristics. The general (GCA) and specific combining ability (SCA) variances were significant (P ≤ 0.01) for all traits studied in the F₁ generation. Three varieties, namely, Sakha108, Giza179, and Sakha109, were highly significantly positive for grain yield plant^-1, which means these varieties can benefit as good combiners to transfer the said trait in the breeding program. In SCA, seven crosses were negative in plant height and a negative value was desirable to avoid lodging and suitable for mechanical harvesting; however, positive significant values are preferable in the rest of the traits. The results indicated that the best crosses for grain yield plant^-1 and plant height were Sakha105 × Sakha102, Sakha105 × Sakha108, and Sakha108 × Sakha109. Conducting cluster analysis also exhibited the genotypes dividing into four groups. The first group included rice genotypes Sakha109 and Sakha108 alone. These cultivars, produced from a common parent Sakha101, could have genetic relations for three quantitative traits (flag leaf area, 1000-grain weight, and panicle weight). The second group, comprising Sakha 102 and Sakha 106, has a very similar genetic background because both cultivars share a common parent, Giza 177. In addition, these two rice cultivars had at par number of tillers plant^-1, 1000-grain weight, and panicles plant^-1. The third group included only Giza 179 belonging to Indica-Japonica varieties. A fourth group has Sakha 105 produced by different parents.

Keywords: rice (Oryza sativa L.), breeding, combining ability, genetic variation, genetic potential, genotypic and phenotypic variances, heritability

Key findings: The half-diallel method was for crossing between six parents of rice to produce 15 crosses. Three parents, namely, Sakha108, Sakha102, and Giza179, were suitable donors for some traits and could serve as better parents for use in breeding programs. Three crosses also gave a good result for grain yield plant^-1 and plant height. PCA and cluster analysis studied the similarity through quantitative traits, providing results that the varieties produced from the same parents cluster in one group.