Multi-environment evaluation of yield stability and adaptability of foxtail millet (Setaria italica (L.) Beauv.) genotypes using AMMI, GGE biplot, and BLUP methods

Breeding foxtail millet for high grain yield and broad agro-climatic adaptability diverse continues to pose a persistent challenge for plant breeders. environments. To meet this objective, 72 genotypes of Foxtail millet were subjected to evaluation through an alpha lattice design in six different environments during two successive rabi seasons (November 2024–March 2025 and November 2025–March 2026) in three Indian locations: Paralakhemundi, Vizianagaram, and Birbhum. The combined ANOVA showed significant effects for genotype, environment, and genotype × environment interaction (GEI). The Additive Main Effects and Multiplicative Interaction (AMMI) analysis indicated that the first two interaction principal component jointly accounted for 76.7% of the GEI variance; notably, genotypes G70(WFx-15), G63(RPCAU-XM-27), G9(GPUF-20), G15(SiA 4253), G54(K2), and G67(IIMR FXM-25 (FX 24042/24062)) were characterized by low Interaction Principal Component Axis scores and exhibited superior stability. The Genotype plus Genotype × Environment biplot analysis grouped all environments into a single mega-environment, identifying E3 and E4 as the most representative environments; notably, G17 (SiA 4389) demonstrated consistent adaptability to all environments and high yield stability. The mean vs. stability graph distinguished genotypes G17(SiA 4389), G31(SiA 3085), G70(WFx-15), G54(K2), G61(KIFXG-22-27), G52(CFXMV-3 (FXV 652/GPUF), and G34(GPUF 19) as high-yielding and stable entries, given their proximity to the Average Environment Coordination line. BLUP-based selection indices, including Harmonic Mean of Genotypic Values (HMGV), Relative Performance of Genotypic Values (RPGV), Harmonic Mean of Relative Performance of Genotypic Values (HMRPGV), and Weighted Average of Absolute Scores and Best Yield (WAASBY), consistently identified G17(SiA 4389), G31(SiA 3085), G70(WFx-15), G54(K2), and G61(KIFXG-22-27) as the promising genotypes combining yield potential with adaptability. Multi-trait selection by MGIDI and MTSI at 15% selection intensity chose 11 of the 72 genotypes each, with strong rank agreement (Spearman’s ρ = 0.792) and six genotypes in common (G52(CFXMV-3 (FXV 652/GPUF), G16(SiA 4216), G34(GPUF 19), G65(BRFoM 4), G41(HMT 100-1), and G15(SiA 4253)). Integration of AMMI stability indices, GGE biplots, BLUP approaches, and multi-trait indices enabled a holistic assessment of genotype performance, allowing for the accurate and robust identification of elite genotypes combining high yield with desirable stability. Based on the combined results of the mean performance and stability analyses, G15 (SiA 4253), G70 (WFx-15) and G54 (K2) were identified as the most desirable genotypes.These findings help identify foxtail millet genotypes with high grain yield and stable performance across the tested environments, providing a basis for their further evaluation in multi-location trials.