Comparative analysis of stability models for identifying rice inter-subspecific breeding lines adapted to different temperature regimes for exploitation in hybrid breeding.

Rice is a staple food crop in Tamil Nadu, cultivated in diverse ecosystems ranging from river delta plains to the Nilgiris hill valleys. The alarming climate change events are predicted to affect rice crop productivity across the globe. In this study, 76 newly developed breeding lines from inter-subspecific crosses, two commercial restorers, and checks were evaluated in four different temperature regimes of Tamil Nadu. Various stability methods were used to analyze genotype-environment interactions to identify lines with stable performance even under various temperature conditions. The stability methods applied in the study were classified under three models viz., the Uni-trait Stability Selection Model (Model - 1), the Uni-trait Mean Performance Stability Selection Model (Model - 2), and the Multi-trait Mean Performance Stability Selection Model (Model - 3). These models are primarily based on Additive Main-effects and Multiplicative Interaction (AMMI), Best Linear Unbiased Prediction (BLUP), and Genotype × Environment (G × E) statistical approaches. Further, molecular markers linked to the Rf3 and Rf4 fertility restorer genes were used to investigate their application in either three-line or two-line hybrid breeding systems. The analysis results revealed a significant genotype-environment interaction in the current study, with temperature being a key factor influencing genotype variation across environments. Various stability models were assessed for efficiency based on correlation and genetic gain results, which indicated that integrating yield performance with stability indices such as GGE (17.51), RPGV (17.51), HMGV (17.51), and WAASBY (166.32) led to higher genetic gain. Furthermore, combining all the models helps to identify lines that are both high-performing and also stable, more effectively than a single model approach. The integrated models identified breeding lines G- 17, G- 25, G- 30, G- 48, and G- 68 as potential candidates for use as restorers in developing hybrids suited to varied-temperature environments, with molecular analysis confirming their use in three-line breeding. Additionally, lines G- 39 and G- 50 are promising candidates for developing climate-smart two-line hybrids with enhanced heterosis.