AQuaRef: machine learning accelerated quantum refinement of protein structures

Cryo-EM and X-ray crystallography provide crucial experimental data for obtaining atomic-detail models of biomacromolecules. Refining these models relies on library-based stereochemical data, which, in addition to being limited to known chemical entities, do not include meaningful noncovalent interactions. Quantum mechanical (QM) calculations could alleviate these issues but are too expensive for large molecules. Here we present a novel AI-enabled Quantum Refinement (AQuaRef) based on AIMNet2 machine learned interatomic potential (MLIP) mimicking QM at substantially lower computational costs. By refining 41 cryo-EM and 30 X-ray structures, we show that this approach yields atomic models with superior geometric quality compared to standard techniques, while maintaining an equal or better fit to experimental data. Notably, AQuaRef aids in determining proton positions, as illustrated in the challenging case of short hydrogen bonds in the parkinsonism-associated human protein DJ-1 and its bacterial homolog YajL.

Authors

Institutions

Publication Details

Journal
Nature Communications
Published
2025-10-17
DOI
https://doi.org/10.1038/s41467-025-64313-1
Citations
8
Primary Topic
RNA and protein synthesis mechanisms
Type
article
Field-Weighted Citation Impact
6.45

Funders

Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
article

AQuaRef: machine learning accelerated quantum refinement of protein structures

Olexandr Isayev, Billy K. Poon, Pavel V. Afonine, Paul D. Adams et al.
8 citations
Nature Communications
RNA and protein synthesis mechanisms
6.45
article

AQuaRef: machine learning accelerated quantum refinement of protein structures

Olexandr Isayev, Billy K. Poon, Pavel V. Afonine, Paul D. Adams, Holger Kruse, Małgorzata Biczysko, Adrián E. Roitberg, Mark P. Waller, Kavindri Ranasinghe, R.I. Zubatyuk, Nigel W. Moriarty, Hatice Gökcan
article en
8 citations

Abstract

Cryo-EM and X-ray crystallography provide crucial experimental data for obtaining atomic-detail models of biomacromolecules. Refining these models relies on library-based stereochemical data, which, in addition to being limited to known chemical entities, do not include meaningful noncovalent interactions. Quantum mechanical (QM) calculations could alleviate these issues but are too expensive for large molecules. Here we present a novel AI-enabled Quantum Refinement (AQuaRef) based on AIMNet2 machine learned interatomic potential (MLIP) mimicking QM at substantially lower computational costs. By refining 41 cryo-EM and 30 X-ray structures, we show that this approach yields atomic models with superior geometric quality compared to standard techniques, while maintaining an equal or better fit to experimental data. Notably, AQuaRef aids in determining proton positions, as illustrated in the challenging case of short hydrogen bonds in the parkinsonism-associated human protein DJ-1 and its bacterial homolog YajL.

Nature CommunicationsVol. 16(1)
WaterNSW (AU), Lawrence Berkeley National Laboratory (US), University of Wrocław (PL), University of Florida (US), Yandex (Russia) (RU), Carnegie Mellon University (US), University of California, Berkeley (US)
National Science Foundation, U.S. Department of Energy, San Diego Supercomputer Center, National Institutes of Health, Office of Science, Office of Advanced Cyberinfrastructure, Division of Chemistry
Openalex Percentile: Top 3%
RNA and protein synthesis mechanisms
6.45
DOIOpen Access PDFOpenAlex
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.