BioMD: All-atom Generative Model for Biomolecular Dynamics Simulation

2025-09-02

AI Summary

Molecular dynamics (MD) simulations are computationally prohibitive for capturing biologically relevant long-timescale processes, while existing machine learning approaches struggle to generate full-atom biomolecular trajectories due to scarcity of long-trajectory data and high training costs. To address this, we propose the first full-atom generative model based on a predictive–interpolative hierarchical framework, enabling efficient synthesis of physically plausible and temporally coherent biomolecular dynamics without requiring long MD trajectories. The model jointly leverages time-series prediction and fine-grained structural interpolation, trained and validated on datasets including DD-13M and MISATO. Experiments demonstrate that generated conformations closely match those from real MD trajectories in quality, and 97.1% of protein–ligand systems successfully yield dissociation pathways within ten sampling attempts. This work substantially enhances the efficiency and feasibility of long-timescale, full-atom biomolecular dynamics modeling.

Technology Category

• Machine Learning: Deep Generative Models & Autoencoders
• Natural Language Processing: Generation
• Cognitive Modeling & Cognitive Systems: Computational Creativity

Application Category

• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Social Networks and Social Media: Generative AI / large language models and their impact on social systems
• User Modeling, Personalization and Recommendation: User modeling and simulation for interactive and conversational systems

Abstract

Molecular dynamics (MD) simulations are essential tools in computational chemistry and drug discovery, offering crucial insights into dynamic molecular behavior. However, their utility is significantly limited by substantial computational costs, which severely restrict accessible timescales for many biologically relevant processes. Despite the encouraging performance of existing machine learning (ML) methods, they struggle to generate extended biomolecular system trajectories, primarily due to the lack of MD datasets and the large computational demands of modeling long historical trajectories. Here, we introduce BioMD, the first all-atom generative model to simulate long-timescale protein-ligand dynamics using a hierarchical framework of forecasting and interpolation. We demonstrate the effectiveness and versatility of BioMD on the DD-13M (ligand unbinding) and MISATO datasets. For both datasets, BioMD generates highly realistic conformations, showing high physical plausibility and low reconstruction errors. Besides, BioMD successfully generates ligand unbinding paths for 97.1% of the protein-ligand systems within ten attempts, demonstrating its ability to explore critical unbinding pathways. Collectively, these results establish BioMD as a tool for simulating complex biomolecular processes, offering broad applicability for computational chemistry and drug discovery.

Problem

• Simulating long-timescale biomolecular dynamics with reduced computational cost
• Overcoming limitations in generating extended trajectories for protein-ligand systems
• Addressing lack of MD datasets and high computational demands

Innovation

• All-atom generative model for biomolecular dynamics
• Hierarchical framework combining forecasting and interpolation
• Generates realistic protein-ligand conformations and unbinding paths