Rapid calculation of accurate atomic charges for proteins via the electronegativity equalization method

Area of Science:

• Computational chemistry
• Biophysics
• Molecular modeling

Background:

• Accurate atomic charges are crucial for molecular simulations of proteins.
• Existing methods for calculating these charges can be computationally expensive.

Purpose of the Study:

• To develop and evaluate empirical models for fast and accurate calculation of conformationally dependent atomic charges in proteins.
• To tailor the electronegativity equalization method (EEM) for protein systems.

Main Methods:

• Parametrization of EEM models using large protein fragments.
• Fitting parameters based on quantum mechanical calculations (Hartree-Fock) with various population analyses (Mulliken, Natural, iterative Hirshfeld) and basis sets (6-31G*, 6-31G**).
• Validation using protein test cases (insulin, ubiquitin) in gas and implicit solvation environments.

Main Results:

• Successfully parametrized and validated 24 EEM models.
• Models demonstrated high accuracy, reproducing quantum mechanics level charges with an average correlation of 0.961, RMSD of 0.097 e, and average absolute error per atom of 0.072 e.
• Results were consistent across different population analyses and basis sets.

Conclusions:

• The developed EEM models provide a fast and accurate method for calculating protein atomic charges.
• These models are suitable for use with the freely available EEM_SOLVER implementation.
• The approach offers a valuable tool for computational studies in biophysics and molecular modeling.