QuantumPDB: A Workflow for High-Throughput Quantum Cluster Model Generation from Protein Structures.

Area of Science:

• Computational chemistry
• Structural biology
• Biophysics

Background:

• Quantum mechanical (QM) calculations offer molecular insights into enzyme catalysis.
• Preparing QM calculations from experimental structures is a bottleneck for high-throughput studies.
• Existing automated tools struggle with diverse active site chemistries and geometries.

Purpose of the Study:

• To develop an automated workflow for generating QM cluster models from raw protein structures.
• To overcome limitations of current automated tools in handling diverse enzyme active sites.
• To enable high-throughput computational screening of enzymes.

Main Methods:

• Developed QuantumPDB, a Python package for automated QM model generation.
• Integrated structure cleaning, protonation state assignment, and QM calculation setup.
• Utilized Voronoi tessellation to create chemically meaningful interaction spheres for QM models.

Main Results:

• Successfully generated 1,673 QM cluster models for 842 holo-enzymes.
• Demonstrated that enzyme environments modulate substrate charge toward neutrality.
• Observed reduction in substrate dipole moment within simulated enzyme active sites.

Conclusions:

• QuantumPDB automates and standardizes multisphere QM model construction.
• Provides a robust platform for large-scale, data-driven protein investigations.
• Facilitates accurate representation of complex active site geometries for QM studies.