Extending the MST Model to Large Biomolecular Systems: Parametrization of the ddCOSMO-MST Continuum Solvation Model

Area of Science:

• Computational chemistry
• Quantum chemistry
• Biophysics

Background:

• Continuum solvation models like PCM and COSMO are essential in quantum chemistry.
• Their application to large biosystems is limited by high computational expense.

Purpose of the Study:

• To parametrize the Miertus-Scrocco-Tomasi (MST) model using the domain decomposition formulation of COSMO (ddCOSMO).
• To enable accurate and cost-effective prediction of hydration free energies for neutral and ionic molecules in large biosystems.

Main Methods:

• Developed the ddCOSMO/MST model, incorporating novel features like hybridization-based atom types and automatic cavity setup for charged regions.
• Parametrized the model at B3LYP/6-31+G(d) and PM6 levels of theory.
• Validated the model on SAMPL2, SAMPL4, and C10 datasets, comparing it with IEFPCM/MST.

Main Results:

• The ddCOSMO/MST model achieves prediction errors of approximately 0.8 kcal/mol for neutral molecules and 3.2 kcal/mol for ions.
• Demonstrated a drastic reduction in computational cost by several orders of magnitude compared to traditional methods.
• Showcased a balanced and accurate description of both cations and anions.

Conclusions:

• The ddCOSMO/MST model offers a significant advancement in computational efficiency and accuracy for solvation free energy calculations.
• This method is robust and suitable for studying large biosystems.
• Provides a reliable tool for predicting hydration free energies of diverse molecular systems.