Toward on-the-fly quantum mechanical/molecular mechanical (QM/MM) docking: development and benchmark of a scoring...

Area of Science:

• Computational chemistry
• Structural biology
• Drug discovery

Background:

• Molecular docking is crucial for drug discovery but faces challenges in accurately modeling polarization and covalent interactions.
• Existing scoring functions often struggle with metalloproteins, particularly those involving zinc ions.

Purpose of the Study:

• To develop and benchmark a novel hybrid quantum mechanical/molecular mechanical (QM/MM) scoring function for molecular docking.
• To improve the accuracy of predicting ligand binding poses in zinc metalloproteins.

Main Methods:

• A QM/MM scoring function was developed using the self-consistent charge density functional tight-binding (SCC-DFTB) method and the CHARMM force field.
• The scoring function was integrated into the EADock DSS docking algorithm.
• A high-quality dataset of X-ray structures for zinc metalloproteins was created and made publicly available for benchmarking.

Main Results:

• The QM/MM scoring function achieved a 77.0% success rate for zinc-bound ligands, a significant improvement over the classical scoring function's 61.5%.
• For allosteric ligands, the success rate remained comparable (49.1%).
• The QM/MM approach markedly improved the identification of correct zinc-binding geometries, boosting docking success rates by over 20% for key drug targets.

Conclusions:

• The developed QM/MM scoring function offers a substantial improvement for docking studies involving zinc metalloproteins.
• This method enhances the prediction of ligand binding poses and geometries, particularly for direct zinc interactions.
• The performance of the QM/MM scoring function is competitive with or superior to widely used docking software.