Scaling Graph Neural Networks to Large Proteins

TL;DR

This paper introduces DISPEF, a large protein dataset, and a new GNN architecture, Schake, to improve the efficiency and transferability of graph neural networks for large biomolecular systems.

Contribution

The study provides a new large-scale protein dataset and a novel multiscale GNN architecture, Schake, enhancing modeling of large proteins with better transferability and efficiency.

Findings

Benchmarking shows importance of long-range interactions in GNNs.

Schake achieves accurate energy and force predictions for large proteins.

DISPEF enables rigorous evaluation of GNNs on complex biomolecular data.

Abstract

Graph neural network (GNN) architectures have emerged as promising force field models, exhibiting high accuracy in predicting complex energies and forces based on atomic identities and Cartesian coordinates. To expand the applicability of GNNs, and machine learning force fields more broadly, optimizing their computational efficiency is critical, especially for large biomolecular systems in classical molecular dynamics simulations. In this study, we address key challenges in existing GNN benchmarks by introducing a dataset, DISPEF, which comprises large, biologically relevant proteins. DISPEF includes 207,454 proteins with sizes up to 12,499 atoms and features diverse chemical environments, spanning folded and disordered regions. The implicit solvation free energies, used as training targets, represent a particularly challenging case due to their many-body nature, providing a stringent test for evaluating the expressiveness of machine learning models. We benchmark the performance of seven GNNs on DISPEF, emphasizing the importance of directly accounting for long-range interactions to enhance model transferability. Additionally, we present a novel multiscale architecture, termed Schake, which delivers transferable and computationally efficient energy and force predictions for large proteins. Our findings offer valuable insights and tools for advancing GNNs in protein modeling applications.