Generalized Born Based Continuous Constant pH Molecular Dynamics in Amber: Implementation, Benchmarking and Analysis.

Area of Science:

• Computational Biology
• Biophysics
• Molecular Dynamics

Background:

• Solution pH significantly impacts biomolecular structure and dynamics.
• Conventional molecular dynamics (MD) struggles to accurately model pH effects due to fixed protonation states.
• Accurate prediction of protein pKa values is essential for understanding biomolecular function.

Purpose of the Study:

• To implement and validate a continuous constant pH molecular dynamics (CpHMD) method using the GBNeck2 generalized Born implicit-solvent model.
• To assess the accuracy and computational cost of the GBNeck2-CpHMD method for predicting protein pKa values.
• To evaluate the method's performance on various biomolecular systems, including proteins with experimentally determined pKa values.

Main Methods:

• CPU-based implementation of CpHMD within the Amber molecular dynamics package, utilizing the GBNeck2 implicit-solvent model.
• pH replica-exchange titration simulations were performed on Asp, Glu, and His side chains in miniproteins and enzymes.
• Comparison of calculated pKa values against experimentally known values, with analysis of computational cost and sampling errors.

Main Results:

• The GBNeck2-CpHMD method achieved a mean unsigned error of 0.70 and a root-mean-squared error of 0.91 in pKa predictions.
• The added computational cost ranged from 11% to 33%, scaling linearly with the ratio of titratable sites to solute atoms.
• The method successfully reproduced experimental pKa orders for coupled carboxylic dyads and identified simulation requirements for buried or interacting residues.

Conclusions:

• The GBNeck2-CpHMD implementation provides an accurate and computationally efficient tool for predicting protein pKa values.
• This method advances the capability of molecular dynamics simulations to account for pH-dependent biomolecular behavior.
• The findings suggest GBNeck2-CpHMD is a valuable asset for researchers studying protein structure, dynamics, and function across various pH conditions.