Identifying Systematic Force Field Errors Using a 3D-RISM Element Counting Correction

Area of Science:

• Computational chemistry
• Molecular modeling
• Physical chemistry

Background:

• Hydration free energies are crucial benchmarks for solvation models.
• Existing models often show errors due to inaccuracies in force fields, not solely solvation models.
• The general AMBER force field (GAFF) requires refinement for accurate predictions.

Purpose of the Study:

• To identify inaccuracies in the general AMBER force field (GAFF) non-bond parameters.
• To evaluate the performance of the 3D reference interaction site model (3D-RISM) with corrections for hydration free energy calculations.
• To propose improvements for force field parameters based on systematic errors.

Main Methods:

• Utilized the 3D reference interaction site model (3D-RISM) for calculating hydration free energies of 642 molecules from the FreeSolv database.
• Applied partial molar volume correction (PMVC) and element count correction (ECC), individually and combined (PMVECC).
• Analyzed systematic errors in parameters for molecules containing specific elements (Cl, Br, I, P).

Main Results:

• The PMVECC approach achieved a mean unsigned error of 1.01±0.04 kcal/mol and RMSE of 1.44±0.07 kcal/mol.
• This performance surpassed benchmark explicit solvent calculations from FreeSolv.
• Identified systematic errors in GAFF parameters for molecules with Cl, Br, I, and P, consistent across methods.

Conclusions:

• The 3D-RISM method with PMVECC offers a computationally efficient and accurate approach for hydration free energy prediction.
• Systematic errors in GAFF parameters, particularly for halogens and phosphorus, are confirmed.
• Minor adjustments to GAFF Lennard-Jones parameters are recommended to enhance hydration free energy predictions for all solvation models.