Dissecting genetic architecture: variance component analysis and heritability estimates for quantitative traits in pearl millet [Pennisetum glaucum (L.) R. Br.]

Abstract

A comprehensive study on the genetic architecture of yield and agronomic traits in pearl millet was conducted using 32 hybrids developed from four CMS lines and eight testers in a Line × Tester design during Kharif 2022 at the National Agricultural Research Project, Ch. Sambhaji Nagar. Variance component analysis was used to estimate additive and dominance effects and heritability for twelve quantitative traits, with the objective of understanding gene action and the relative contribution of lines, testers, and their interactions. Analysis of variance revealed highly significant genotypic differences for all traits except downy mildew resistance, indicating substantial genetic variability. Grain yield was predominantly governed by additive gene action (GCA = 0.10, SCA = 0.03; A:D = 3.33) with high narrow-sense heritability (83.02%), suggesting effectiveness of pedigree or recurrent selection. Days to 50% flowering showed nearly equal additive and dominance effects (A:D = 1.01; h² = 63.09%), allowing both parent selection and heterosis breeding. In contrast, plant height, ear-head length, and zinc content were entirely controlled by dominance effects with negligible additive variance (h² = 0.00%). Traits such as days to maturity, productive tillers, ear-head girth, 1000-seed weight, fodder yield, and iron content exhibited non-additive gene action with low to moderate heritability (12.71-41.76%), highlighting the importance of hybrid evaluation to exploit dominance and epistasis. Variance partitioning showed that testers contributed most to grain yield (79.87%), days to 50% flowering (59.64%), and ear-head length (54.00%), reflecting strong additive effects from male parents. Line × Tester interactions were predominant for plant height (76.21%), productive tillers (78.07%), and zinc content (81.86%), indicating the importance of specific parental combinations. Overall, the results emphasize trait-specific breeding strategies: additive traits favor parent selection and pedigree breeding, while non-additive traits require focused hybrid screening.