Stable high-yield pearl millet genotypes identified from global germplasm through AMMI, GGE, and stability index approaches

Context Pearl millet, a climate-resilient, nutrient-rich cereal, faces productivity challenges from environmental variability, necessitating identification of stable, high-yielding genotypes. Aims This study aimed to identify stable, high-yielding pearl millet to achieve increased grain yield per hectare (GYPH), spike length (SL), spike girth (SG), and thousand-seed weight (TSW) from a global germplasm collection of 248 pearl millet genotypes. Methods Field trials were conducted in an alpha lattice design with three replications across three environments. Genotype stability and performance were assessed using additive main effect and multiplicative interaction (AMMI) and genotype main effect plus genotype-by-environment interaction (GGE) biplot analyses. Stability indices (AMMI-based stability parameter (ASTAB), AMMI stability index (ASI), AMMI stability value (ASV), modified AMMI stability index (MASI), and modified AMMI stability value (MASV)) were integrated to quantitatively assess the stable genotypes. Key results The first two AMMI principal components explained 55% and 45% of the total variation for GYPH, whereas GGE biplots explained 84.99%, 88.89%, 77.7%, and 83.17% for GYPH, SL, SG, and TSW respectively. AMMI identified G57, G101, and G209 as stable for GYPH, whereas GGE selected G87, G242, G246, and G131. The genotype selection index (GSI) highlighted G87, G209, G242, G143, and G172 as highly stable for GYPH. Stable genotypes were also identified for SL (G57, G246, G140), SG (G218, G157, G8), and TSW (G57, G95, G149). Conclusions The integrated assessment of stability using multiple approaches was shown to be effective in identifying stable genotypes across diverse environments. Notably, G87 emerged as the most stable genotype for GYPH. Implications Integrating stable genotypes into breeding programs could enhance yield stability, disease resistance, and grain quality, ensuring adaptability across environments.