Multi agent AI system improves drug molecule design success rates
M3OS: A Monte Carlo Graph Search-Orchestrated Multi-Agent LLM System for Evidence-Traced Molecular Optimization
Machine Learning
Summary
Designing new drug molecules is a complex task that needs careful reasoning about chemistry and previous test results. The authors created M3OS, a system where many AI agents work together and keep track of past molecule designs and their evaluations using a graph. This helps the system remember what worked before and make better decisions about which molecules to try next. Their approach performed better on several molecular design problems compared to other methods.
What this means in practice
- •For drug discovery teams: Design drug molecules better by integrating chemical reasoning and test results in an AI-driven iterative search process.
- •For software engineers in chemistry: Build multi-agent systems that coordinate AI models through persistent graphs to manage complex design workflows.
Authors
Junjie Wang, Yaowei Jin, Ruohui Tang, Guonan Cui, Haojie Wang, Penglei Wang, Dingyan Wang, Duo An, Shuangjia Zheng, Qian Shi
Abstract
Small-molecule optimization integrates medicinal-chemistry reasoning and computational evidence through iterative, multi-objective decisions. When large language models (LLMs) reason over optimization histories stored primarily in conversational context, they must recover candidate identities, prior evaluations, and task constraints to guide subsequent decisions. We present M3OS, a multi-agent LLM system that decouples molecular-design reasoning from optimization-state management through Monte Carlo graph search. A persistent graph links evaluated candidates, parent-child transformations and evaluation evidence, while rewards and visit statistics guide LLM-assisted parent selection. Two branches combine tool-driven candidate generation with knowledge- and case-guided medicinal-chemistry editing. An execution harness controls graph updates through structured output extraction, molecular validation and task-bound evaluation. Agents receive role-specific contexts, while the graph preserves optimization trajectories beyond their active contexts. Across three molecular optimization benchmarks, M3OS achieves higher success rates than baselines, supporting the integration of persistent search state, specialized agents and controlled execution for multi-constraint optimization.