Identifying Drought-Tolerant Rice Genotypes using Drought Response Indices and Machine Learning

• Published in: 2024 International Conference on Cognitive Robotics and Intelligent Systems (ICC - ROBINS) pp. 305 - 311
• Main Authors: Daniels, D Alan, Binodh, Asish K., Raimond, Kumudha, Thankappan, Sugitha
• Format: Conference Proceeding
• Language: English
• Published: IEEE 17-04-2024
• Subjects: Analytical models; climate change; Cognitive robotics; Correlation; Drought tolerance; Droughts; food security; Intelligent systems; Machine learning; phenotypic traits; Predictive models; rice genotypes; water scarcity
• DOI: 10.1109/ICC-ROBINS60238.2024.10533898

This study has investigated the development of a predictive model for drought tolerance in rice genotypes, leveraging their phenotypic traits. A dataset encompassing thirteen phenotypic parameters measured in hundred rice genotypes under both stressed and non-stressed conditions has been utilized. Drought Response Indices (DRIs) have been calculated for each parameter and genotype, additionally introducing a novel "Total DRI" to represent the overall drought response. Through correlation analysis and multiple regression techniques, key traits associated with drought tolerance have been identified and employed to build robust models for predicting "Total DRI". The selected model's performance has been evaluated by its ability to predict the "Total DRI" of reference genotypes representing drought susceptibility and tolerance. Furthermore, the chosen model has been applied to new rice genotypes for screening and selection of drought-tolerant varieties. This research contributes to the advancement of resilient rice cultivars suitable for arid regions.