Rice production is currently challenged by high greenhouse gas emissions, low nitrogen uses efficiency (NUE), and losses from major diseases. Solving these together requires solutions that cut emissions and fertilizer use while maintaining productivity.

These challenges are closely tied to what happens around plant roots, where microbes shape methane emissions, nutrient cycling, and plant health. But using single “beneficial” microbes has often given mixed results because they struggle to survive and perform consistently in different environments.

AI-SynCom-Rice takes a different approach by designing Synthetic Communities (SynComs), which are carefully selected groups of microbes that work together rather than alone. Artificial intelligence and machine learning help analyze large microbiome datasets to find combinations that are stable and effective.

The approach builds on IRRI’s long-term rice–upland crop rotation experiment to identify locally adapted microbes linked to lower emissions, better nitrogen efficiency, and stronger natural nitrogen fixation. These insights are combined with metagenomics, predictive modeling, and large-scale screening to refine microbial combinations.

The goal is to develop an AI-guided pipeline that can design microbial communities to reduce methane emissions, improve nitrogen efficiency, and strengthen disease resistance, supporting more sustainable and climate-smart rice systems.

Supported by CGIAR’s Sustainable Farming Initiative (AoW4), with partner expertise from the Japan International Research Center for Agricultural Sciences (JIRCAS), the University of Bremen, and Cornell University.