MDForge: Agentic Molecular Dynamics Pipeline Design under Sparse Simulator Feedback

2026-06-11

AI Summary

This work addresses the high reliance on expert knowledge and costly trial-and-error in molecular dynamics (MD) simulation pipeline design. The authors propose a large language model (LLM)-based multi-agent system that frames MD pipeline construction as an open-ended code generation task. For the first time, they introduce a multi-agent debate mechanism coupled with verbal rewards to enable autonomous pipeline optimization under sparse feedback, moving beyond conventional predefined tool-calling paradigms. Evaluated on three SAMPL benchmarks, the method automatically generates MD workflows comparable to those crafted by human experts and successfully identifies a novel cucurbit[7]uril (CB[7]) binder with picomolar affinity, later validated by NMR spectroscopy.

Abstract

Molecular dynamics (MD) is the canonical in-silico method for atomistic molecular science, simulating molecular behavior from first-principle physics. Designing an MD pipeline for a new system requires substantial expert knowledge: running it on even one molecule is expensive, ruling out trial-and-error. We automate this expert pipeline-design process with an LLM agent. Unlike existing MD agents that orchestrate a predefined tool set, we treat pipeline design as open-ended code generation in which the agent's behavior is reshaped online by verbal reward. Specifically, we build MDForge, an LLM agent whose in-context update rule densifies the sparse reward via a multi-agent debate among physics experts. On three SAMPL host-guest binding free-energy benchmarks, MDForge automatically designs MD pipelines competitive with human experts. Deployed on a library of unseen candidate guests, its CB[7] pipeline discovers a novel binder that wet-lab competition NMR confirms is a high-affinity, picomolar CB[7] binder. Our data and code are available at https://github.com/Zehong-Wang/MDForge.

Problem

• Molecular Dynamics
• Pipeline Design
• Sparse Feedback
• Automated Simulation
• Binding Free Energy

Innovation

• LLM agent
• molecular dynamics
• sparse reward densification
• multi-agent debate
• automated pipeline design