MD-LLM-1: A Large Language Model for Molecular Dynamics

TL;DR

This paper introduces MD-LLM-1, a large language model fine-tuned on protein conformations, capable of predicting unseen states and exploring protein conformational landscapes, thus offering a new approach to molecular dynamics modeling.

Contribution

The paper presents the first implementation of a large language model for molecular dynamics, fine-tuned on protein data to predict multiple conformational states.

Findings

MD-LLM-1 can predict unseen conformational states of proteins.

Training on one state enables exploration of other states.

The model learns principles of protein conformational landscapes.

Abstract

Molecular dynamics (MD) is a powerful approach for modelling molecular systems, but it remains computationally intensive on spatial and time scales of many macromolecular systems of biological interest. To explore the opportunities offered by deep learning to address this problem, we introduce a Molecular Dynamics Large Language Model (MD-LLM) framework to illustrate how LLMs can be leveraged to learn protein dynamics and discover states not seen in training. By applying MD-LLM-1, the first implementation of this approach, obtained by fine-tuning Mistral 7B, to the T4 lysozyme and Mad2 protein systems, we show that training on one conformational state enables the prediction of other conformational states. These results indicate that MD-LLM-1 can learn the principles for the exploration of the conformational landscapes of proteins, although it is not yet modeling explicitly their thermodynamics and kinetics.