Geometric learning of the conformational dynamics of molecules using dynamic graph neural networks

TL;DR

This paper introduces a dynamic graph neural network approach to predict the conformational changes of molecules over time, achieving chemically accurate distance predictions and demonstrating transferability across different molecular systems.

Contribution

The study presents a novel application of temporal edge prediction in dynamic graphs for molecular conformational dynamics, with high accuracy and transferability.

Findings

Mean absolute error of 0.017 Å, considered chemically accurate.

Transfer learning reduces training data needs significantly.

Model generalizes well to new molecular systems with minimal fine-tuning.

Abstract

We apply a temporal edge prediction model for weighted dynamic graphs to predict time-dependent changes in molecular structure. Each molecule is represented as a complete graph in which each atom is a vertex and all vertex pairs are connected by an edge weighted by the Euclidean distance between atom pairs. We ingest a sequence of complete molecular graphs into a dynamic graph neural network (GNN) to predict the graph at the next time step. Our dynamic GNN predicts atom-to-atom distances with a mean absolute error of 0.017 \r{A}, which is considered ``chemically accurate'' for molecular simulations. We also explored the transferability of a trained network to new molecular systems and found that finetuning with less than 10% of the total trajectory provides a mean absolute error of the same order of magnitude as that when training from scratch on the full molecular trajectory.