All-atomistic Transferable Neural Potentials for Protein Solvation

TL;DR

This paper introduces PHNN, a neural network-based implicit solvent model for proteins that improves accuracy and transferability by learning corrections to model parameters, leveraging physical priors, and maintaining performance on diverse systems.

Contribution

The paper presents PHNN, a novel neural potential that enhances implicit solvent modeling for proteins by learning transferable corrections, addressing transferability and accuracy issues.

Findings

PHNN outperforms traditional analytical methods in accuracy.

PHNN maintains predictive accuracy on out-of-domain protein systems.

The model leverages physical priors for data efficiency.

Abstract

Implicit solvent models are widely used to decrease the number of solvent degrees of freedom and enable the calculation of solvation energetics without water molecules. However, its accuracy often falls short compared to explicit models. Recent advancements in neural potentials have shown promise in drug discovery, but transferability remains a persistent challenge. Here, we introduce the Protein Hydration Neural Network (PHNN), an implicit solvent model that extends analytical continuum solvation by learning transferable corrections to model parameters instead of applying post hoc adjustments to final energies. The model is explicitly designed to maximize data efficiency by leveraging physical priors embedded in the data. We demonstrate that PHNN improves accuracy relative to traditional analytical methods and maintains predictive accuracy on out-of-domain protein systems.