Saved in:
Bibliographic Details
Main Authors: Zeng, Lejia, Zhang, Xintong, Pei, Yuchan, Zhao, Lifeng, Hua, Lan, Yang, Jincai, Huang, Niu
Format: Preprint
Published: 2026
Subjects:
Online Access:https://arxiv.org/abs/2601.11628
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866915757319782400
author Zeng, Lejia
Zhang, Xintong
Pei, Yuchan
Zhao, Lifeng
Hua, Lan
Yang, Jincai
Huang, Niu
author_facet Zeng, Lejia
Zhang, Xintong
Pei, Yuchan
Zhao, Lifeng
Hua, Lan
Yang, Jincai
Huang, Niu
contents Machine learning interatomic potentials (MLIPs) enable efficient modeling of molecular interactions with quantum mechanical (QM) accuracy. However, constructing robust and representative training datasets that capture subtle, system-specific interaction motifs remains challenging. We introduce PANIP (PAirwise Non-covalent Interaction Potential), an ensemble MLIP model built upon the NequIP framework and trained on non-covalent interactions (NCIs) between protein-derived fragments. PANIP is trained using an automated multi-fidelity active learning (MFAL) workflow, in which a representative training subset, termed PDB-FRAGID (PDB Fragment Interaction Dataset), was distilled from an otherwise prohibitively large pool of fragment dimers extracted from the Protein Data Bank (PDB). PANIP retains $ω$B97X-D3BJ/def2-TZVPP-level accuracy and achieves mean absolute errors below 0.2 kcal/mol on out-of-distribution systems, demonstrating excellent transferability across diverse NCI motifs. Compared to the widely used ANI-2x potential, PANIP delivers substantially lower errors, particularly for charged and strongly interacting dimers. Coupled with a fragmentation-based energy decomposition scheme, PANIP estimates protein-ligand binding energies at near force-field computational cost yet QM-level accuracy, enabling its use as a fragment-based scoring function that rivals specialized docking scoring functions.
format Preprint
id arxiv_https___arxiv_org_abs_2601_11628
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Developing a Machine-Learning Interatomic Potential for Non-Covalent Interactions in Proteins
Zeng, Lejia
Zhang, Xintong
Pei, Yuchan
Zhao, Lifeng
Hua, Lan
Yang, Jincai
Huang, Niu
Chemical Physics
Machine learning interatomic potentials (MLIPs) enable efficient modeling of molecular interactions with quantum mechanical (QM) accuracy. However, constructing robust and representative training datasets that capture subtle, system-specific interaction motifs remains challenging. We introduce PANIP (PAirwise Non-covalent Interaction Potential), an ensemble MLIP model built upon the NequIP framework and trained on non-covalent interactions (NCIs) between protein-derived fragments. PANIP is trained using an automated multi-fidelity active learning (MFAL) workflow, in which a representative training subset, termed PDB-FRAGID (PDB Fragment Interaction Dataset), was distilled from an otherwise prohibitively large pool of fragment dimers extracted from the Protein Data Bank (PDB). PANIP retains $ω$B97X-D3BJ/def2-TZVPP-level accuracy and achieves mean absolute errors below 0.2 kcal/mol on out-of-distribution systems, demonstrating excellent transferability across diverse NCI motifs. Compared to the widely used ANI-2x potential, PANIP delivers substantially lower errors, particularly for charged and strongly interacting dimers. Coupled with a fragmentation-based energy decomposition scheme, PANIP estimates protein-ligand binding energies at near force-field computational cost yet QM-level accuracy, enabling its use as a fragment-based scoring function that rivals specialized docking scoring functions.
title Developing a Machine-Learning Interatomic Potential for Non-Covalent Interactions in Proteins
topic Chemical Physics
url https://arxiv.org/abs/2601.11628