Back

ChatGEM: An Agentic Architecture Enabling Interactive Simulation of Genome-Scale Metabolic Models

Chowdhury, N.; George, A.; Purohit, S.; Contolesi, A.; Bredeweg, E. L.; Czajka, J.; Stratton, K. G.; Gao, Y.; Stephenson, M.; Elmore, J. R.; Scott, A.; Leach, D. T.; Jerger, A.; Lemmon, T.; Piehowski, P.; Tate, K.; Fulcher, J. M.; Beliaev, A.; Burnum-Johnson, K.; Rigor, P.; Bardhan, J.

2026-07-21 systems biology
10.64898/2026.07.20.739662 bioRxiv
Show abstract

Genome-scale metabolic models (GEMs) are powerful tools for predicting cellular phenotypes and guiding microbial strain engineering, yet broad adoption remains challenging due to the computational expertise required. To overcome that, we present ChatGEM, an agentic platform that enables interactive GEM simulation through natural language. Built on the multi-agent ADEPT framework, ChatGEM integrates COBRApy within a retrieval-augmented generation (RAG) architecture that coordinates code generation and execution through specialized agents. Benchmarking across three tasks of increasing complexity showed that RAG-enabled code generation improved the mean overall performance score from 2.63 to 4.20 while reducing the execution time significantly starting from routine to complex tasks. Application of ChatGEM using an enzyme-constrained GEM (ecGEM) for four engineered Pseudomonas putida KT2440 strains identified the constitutive strain as the optimal chassis for succinate overproduction using a succinate leakage index - a prediction observed experimentally. Therefore, ChatGEM democratizes metabolic modeling by enabling researchers without computational expertise to perform sophisticated GEM-based analyses through natural language, and, hence, accelerating scientific discovery.

Matching journals

The top 3 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.