Identifying Key Selection Criteria for Grain Yield and Starch Content Improvement in Doubled Haploid lines of maize

Abstract

The present study assessed genetic variability and trait associations to identify potential selection criteria for improving grain yield and starch content in doubled haploid (DH) maize lines. Ninety DH lines and four replicated checks were evaluated during kharif, 2025 using an augmented randomized complete block design at the Maize Research Unit, Agricultural Research Institute, Rajendranagar, Hyderabad, Telangana. Adjusted means were used for genetic variability, correlation and path coefficient analyses. Significant treatment variation was observed for most traits. Grain yield exhibited high broad-sense heritability (96.35%) and genetic advance as a percentage of mean (26.24%), indicating substantial potential for selection. Grain yield showed strong positive correlations with starch content, number of kernels per row, ear diameter and ear length. Before path analysis, variance inflation factor values for the seven selected predictor traits ranged from 1.59 to 4.26, indicating no severe multicollinearity. Starch content exhibited the highest positive direct effect on grain yield (0.644), followed by number of kernels per row (0.253) and ear diameter (0.129). The seven-trait model explained 92.45% of the variation in adjusted mean grain yield. Notably, DH-62 recorded the highest grain yield (108.03 g plant–¹) and starch content (63.79%). These findings identify starch content, number of kernels per row and ear diameter as promising selection attributes for improving grain yield, with potential for simultaneous improvement of productivity and starch quality. Further validation across environments and combining-ability studies are required before utilizing promising DH lines in hybrid development.