Adding the AMBER 14SB Force Field to the Stochastic Titration CpHMD Method.

Area of Science:

• Computational chemistry
• Biomolecular simulations
• Protein dynamics

Background:

• Accurate molecular dynamics (MD) simulations require incorporating pH effects to capture conformational, energetic, and protonation states.
• Constant-pH molecular dynamics (CpHMD) methods, particularly stochastic titration CpHMD (st-CpHMD), are advanced techniques for pH-dependent simulations.
• st-CpHMD currently supports GROMOS 54A7 and CHARMM 36m force fields.

Purpose of the Study:

• To extend stochastic titration CpHMD (st-CpHMD) compatibility to the AMBER 14SB force field within the GROMACS software package.
• To validate the modified AMBER 14SB force field for st-CpHMD simulations.
• To enable direct performance comparisons between GROMOS 54A7, CHARMM 36m, and AMBER 14SB force fields in CpHMD.

Main Methods:

• Implemented st-CpHMD support for the AMBER 14SB force field in GROMACS.
• Introduced a minor modification to AMBER 14SB atomic partial charges for main chain neutralization.
• Benchmarked the implementation using lysozyme and Staphylococcal nuclease proteins, comparing pKa predictions and computational costs.

Main Results:

• The AMBER 14SB implementation achieved pKa prediction accuracy comparable to other force fields, with low root-mean-square error (RMSE).
• AMBER 14SB simulations exhibited lower computational cost than CHARMM 36m but higher than GROMOS 54A7.
• Identified challenging cases requiring further method refinement and suggested accelerating the PB/MC step for improved computational speed.

Conclusions:

• Developed the first CpHMD method compatible with GROMOS 54A7, CHARMM 36m, and AMBER 14SB force fields.
• Enabled direct, quantitative performance comparisons of major protein force fields for CpHMD simulations.
• Highlighted the need for further optimization, particularly in the PB/MC step, to enhance computational efficiency.