Performance of doubled haploid maize (Zea mays L.) testcross hybrids under optimal and drought-stressed environments

Maize (Zea mays L.) productivity in Sub-Saharan Africa is increasingly constrained by recurrent drought linked to climate change. Improving yield stability under contrasting moisture conditions remains challenging, especially when breeding materials are derived from parental lines within the same heterotic group (e.g., Group A × Group A), where genetic divergence is limited. We hypothesized that doubled haploid (DH) lines derived from biparental populations still harbor sufficient within-population genetic variation to generate exploitable phenotypic diversity and adaptive differentiation. Furthermore, crossing each DH line with a single-cross tester from the opposite heterotic group provides an effective framework to capture this variation through hybrid performance. In the present work therefore, 855 DH testcross hybrids and six commercial checks were evaluated under optimal and managed drought conditions using an alpha-lattice design. Drought stress was imposed two weeks before flowering until harvest. The mean grain yield under optimal conditions were between 3.66 to 10.36 t ha-1, while 0.16 to 6.13 t ha-1 under drought condition. The mean of top 10 hybrids (≈1.2% selection intensity) outperformed the mean of commercial checks by 60% under optimal conditions and 161% under drought stress. Under optimum conditions, 629 and 286 hybrids exhibited higher heterosis over the mean of checks and the best check, respectively. Similarly, 536 and 235 hybrids surpassed the mean of checks and the best check, respectively, in terms of standard heterosis. The mean and best check yields under drought stress were 1.97 and 3.10 while, 5.91 and 7.09 t ha-1 under optimum condition, respectively. The multi-trait clustering grouped the hybrids into four distinct adaptive categories. Cluster 3, defined primarily by grain yield, integrated temperate introgression with elite tropical backgrounds consistently expressed superior yield performance along with desirable secondary traits. These results demonstrate that DH lines derived from within-group parental crosses can generate functional diversity and predictable adaptive clusters. Thus supports a cluster-based selection strategy to improve drought tolerance, yield potential, and adaptation in maize breeding programs.