LASI: Quantum Mechanics/Molecular Mechanics Data for Enzymatic Phosphate Hydrolysis

Dénes Berta's dataset, published on figshare in March 2026, contains computational data for modeling enzyme-catalyzed phosphate hydrolysis. The data supports the Locality Accelerated and Systematically Improvable (LASI) protocol, which combines quantum embedding and local natural orbital methods. It is designed to achieve chemical accuracy at an affordable computational cost for large quantum subsystems.

Use Cases

• Benchmarking density functional theory (DFT) models against CCSD(T) reference energetics based on the described methodology.
• Training or validating machine learning models for predicting enzymatic reaction barriers using large QM subsystem data.
• Studying the effects of quantum subsystem size (up to ~400 atoms) on calculated reaction energies and barriers.
• Developing quantum embedding protocols for cost reduction in large-scale QM and QM/MM simulations.

Strengths

• Data is associated with a protocol validated on clinically relevant enzyme-catalyzed phosphate hydrolysis.
• The methodology is designed for large quantum subsystems up to approximately 400 atoms.
• Dataset is published under a specific CC-BY-NC-4.0 license, clarifying reuse terms.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count is unknown, which may limit suitability assessment.
• The dataset size is 1.5 MB, suggesting it may contain summary or model data rather than extensive raw simulation outputs.

Provenance

Source

Dénes Berta via figshare.

Collection Method

Data generated through computational simulations using the proposed LASI QM/MM protocol.

Time Range

Publication date is 2026-03-25; temporal coverage of the underlying simulations is not specified.

Freshness

Last updated 2026-03-25 02:03:37.

Geography

Spatial coverage is not applicable for computational chemistry data.