Development and Comprehensive Benchmark of a High-Quality AMBER-Consistent Small Molecule Force Field with Broad...

Area of Science:

• Computational chemistry
• Molecular modeling
• Drug discovery

Background:

• Molecular mechanics force fields (FFs) are crucial for biomolecular simulations in drug discovery.
• Developing high-quality, general FFs is resource-intensive and often limited to a few organizations.
• Existing FFs may have limited training sets, simplified representations, or restricted access.

Purpose of the Study:

• To develop an AMBER-consistent small molecule FF with broad chemical space coverage.
• To provide open-access parameters for the molecular modeling community.
• To validate the new FF's performance against established methods.

Main Methods:

• Quantum mechanics (QM)/molecular mechanics (MM) conformer comparison benchmarks.
• Free energy perturbation (FEP) calculations on benchmark datasets.
• Validation against QM energies, geometries, and experimental binding free energies.

Main Results:

• The new FF outperforms OpenFF2 and GAFF2 in reproducing QM energies and geometries.
• Achieved root-mean-square errors of 1.19 kcal/mol for ΔΔG and 0.92 kcal/mol for ΔG in relative binding free energy calculations.
• Performance is comparable to leading commercial force fields like OPLS.

Conclusions:

• The developed open-access FF offers a high-quality, broadly applicable alternative for biomolecular simulations.
• It enhances the accuracy and accessibility of computational drug discovery tools.
• The FF provides reliable predictions for molecular interactions and binding affinities.