All-atomistic Transferable Neural Potentials for Protein Solvation

2026-05-14

AI Summary

Traditional implicit solvent models exhibit limited accuracy in predicting protein solvation energies, while existing neural potentials often lack transferability across diverse systems. To address these challenges, this work proposes the Protein Hydration Neural Network (PHNN), which integrates physical priors into a data-driven framework to enable end-to-end learning of transferable corrections to continuum solvent model parameters—rather than applying post hoc energy adjustments. By directly refining the underlying solvation model within a physically informed architecture, PHNN achieves substantially improved prediction accuracy and demonstrates strong generalization performance even on out-of-distribution protein systems, thereby overcoming a key transferability bottleneck that has hindered the application of neural potentials in implicit solvent modeling.

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.

Problem

• implicit solvent model
• solvation energetics
• transferability
• neural potentials
• protein solvation

Innovation

• neural potentials
• implicit solvent model
• transferability
• protein solvation
• physical priors