Fast empirical pKa prediction by Ewald summation

Area of Science:

• Computational Chemistry
• Biophysics
• Structural Biology

Background:

• Macromolecule pK(a) calculations typically use the Poisson-Boltzmann equation.
• Existing methods are too slow for high-throughput analysis or periodic boundary conditions in molecular dynamics simulations.

Purpose of the Study:

• Develop a rapid and accurate pK(a) calculation method suitable for large-scale analysis and periodic systems.
• Address limitations of current Poisson-Boltzmann solvers in specific computational contexts.

Main Methods:

• An empirical equation relating pK(a) to electrostatic potential, hydrogen bonds, and accessible surface area was developed.
• Electrostatic potential was calculated using Ewald summation with Gaussian charge densities.
• Empirical constants were derived from 217 experimentally determined pK(a) values.

Main Results:

• The empirical method achieves high overall jack-knifed accuracy.
• The calculation speed is sufficient for real-time monitoring during molecular dynamics simulations.
• The approach is effective for predicting pK(a)s in periodic crystal environments.

Conclusions:

• The proposed empirical equation offers a fast and accurate alternative for pK(a) calculations.
• This method enhances the feasibility of large-scale pK(a) analysis and in silico crystal simulations.
• A reliable null-model for assessing pK(a) prediction accuracy was also presented.