Year: 2026
Type: preprint
Abstract: We introduce QMzyme, a Python package to automate the generation of Quantum mechanics (QM)-based enzyme models. QM calculations on enzymes are computationally challenging because full enzyme systems typically contain thousands of atoms, placing accurate QM treatments beyond practical computational limits.Therefore, QM-based enzyme studies rely on identifying a subset of atoms that captures the essential chemistry of the reaction environment, enabling tractable yet chemically meaningful calculations. However, constructing these reduced models requires numerous user-defined decisions, including selection of the starting structure, identification of atoms to include in the QM region, and definition of truncation strategies for QM-only models, all of which can substantially influence the resulting calculations. QMzyme aims to provide the community with a flexible and extensible framework for automated, systematic generation of QM-based enzyme models, enabling researchers to more rigorously evaluate how model construction decisions influence computational outcomes while improving reproducibility and transparency across workflows. By facilitating high-throughput generation and analysis of QM enzyme models, QMzyme supports largescale, data-driven investigations of enzyme structure, function, and reactivity. This effort is especially crucial for the creation of large, chemically consistent datasets that can support emergent artificial intelligence and machine learning approaches for accelerated digital discovery in enzymology and biocatalysis.
Source: ChemRxiv
Authors: Arkanil Roy, Young Woo Kim, Heidi Klem
Institution: Louisiana State University
Language: English
Subfield: Atomic and Molecular Physics, and Optics
Field: Physics and Astronomy
Domain: Physical Sciences
Open Access status: green
Awards: G1117273162, G1453181110, Track 4: Advanced CI Coordination Ecosystem: Monitoring and Measurement Services, G3062527969, G3562386289, G3778588758, Track 1: ACCESS Resource Allocations Marketplace and Platform Services (RAMPS), G4010157413, ACO: An Open CI Ecosystem to Advance Scientific Discovery (OpenCI), Track 3: COre National Ecosystem for CyberinfrasTructure (CONECT), G5367127336, G5464919783, G5483133893, G8054376542, Track 2: Customized Multi-tier Assistance, Training, and Computational Help (MATCH) for End User ACCESS to CI
OpenAlex ID: W7215083574