A multi-model genotype × environment interaction analysis discerning phenotypic plasticity of the strong culm trait in rice

A strong culm is a key trait for reducing lodging in rice, arising from the intricate coaction of morphological and environmental factors. As both grain yield (GY) and culm strength (CS) are complex traits, modeling genotype × environment interaction (G × E) and quantifying genotypic stability are essential for recommending genotypes in multi-environment trials (METs). We evaluated a selected set of recombinant inbred lines (RILs) derived from indica cv. Swarna and tropical japonica acc. IRGC 39111, along with checks across transplanted and direct-seeded environments, employing a multi-model approach to dissect mean performance and stability (MPS) for CS and GY traits. Phenotypic plasticity and stability were assessed using parametric and multivariate approaches, emphasizing AMMI, GGE, BLUPs, and the multi-trait stability index (MTSI). We further computed the weighted average of absolute scores (WAASB) to summarize G × E effects and WAASBY to integrate MPS. Selection was guided by genotype–ideotype distance via MTSI, and a genotype selection index (GSI) was used to synthesize yield and stability rankings. Breaking resistance (BR) was used as the primary proxy for CS. Significant G × E effects were detected for all traits. WAASB and WAASBY effectively captured stability and MPS, enabling the discrimination of broadly adapted and high-performing RILs. Multi-trait selection via MTSI improved selection efficiency relative to single-trait criteria and was consistent with AMMI-, GGE-, and BLUP-based inferences. The GSI identified RIL 315 (G16) as superior for GY and RIL 421 (G3) for BR. Consistently across AMMI, GGE, and MTSI, RIL 417 (G1), RIL 419 (G2), RIL 314 (G15), and tropical japonica acc. IRGC 10658 exhibited stable and high GY coupled with a strong culm across environments. A multi-model, multi-trait strategy robustly identified strong-culm, high-yielding RILs across variable environments. The convergence of WAASB/WAASBY, MTSI, and GSI supports the confident advancement of candidate lines and broadens the phenotyping framework and germplasm base for lodging resistance in future breeding programs.