Ligand-induced protein dynamics differences correlate with protein-ligand binding affinities: An unsupervised deep learning approach

TL;DR

This paper introduces an unsupervised deep learning approach that analyzes protein dynamics to predict protein-ligand binding affinities, revealing how ligand-induced dynamic changes relate to binding strength and identifying key residues involved.

Contribution

The study presents a novel unsupervised deep learning method that correlates protein dynamics with binding affinity and identifies critical residues, advancing dynamics-based drug discovery.

Findings

Dynamic features strongly correlate with binding affinities

Residues important for interactions are identified based on their contribution

Method demonstrates potential for predicting binding affinities from simulation data

Abstract

Prediction of protein-ligand binding affinity is a major goal in drug discovery. Generally, free energy gap is calculated between two states (e.g., ligand binding and unbinding). The energy gap implicitly includes the effects of changes in protein dynamics induced by the binding ligand. However, the relationship between protein dynamics and binding affinity remains unclear. Here, we propose a novel method that represents protein behavioral change upon ligand binding with a simple feature that can be used to predict protein-ligand affinity. From unbiased molecular simulation data, an unsupervised deep learning method measures the differences in protein dynamics at a ligand-binding site depending on the bound ligands. A dimension-reduction method extracts a dynamic feature that is strongly correlated to the binding affinities. Moreover, the residues that play important roles in protein-ligand interactions are specified based on their contribution to the differences. These results indicate the potential for dynamics-based drug discovery.