Performance evaluation of the zero-multipole summation method in modern molecular dynamics software.

Area of Science:

• Computational Chemistry
• Molecular Dynamics Simulations
• Biophysics

Background:

• Accurate calculation of electrostatic interactions is crucial for molecular dynamics (MD) simulations.
• The zero-multiple summation method (ZMM) is a cutoff-based approach for electrostatic calculations.
• Previous studies confirmed the accuracy of ZMM, necessitating performance evaluation.

Purpose of the Study:

• To compare the practical performance of the zero-multiple summation method (ZMM) against the smooth particle mesh Ewald method (SPME).
• To evaluate ZMM's performance in molecular dynamics simulations using the GROMACS software package.
• To analyze the impact of potential parameters and CPU core count on ZMM and SPME performance.

Main Methods:

• Implementation of ZMM and SPME within the GROMACS MD simulation software.
• Performance benchmarking using various-sized water systems and protein-water systems.
• Detailed analysis of performance dependency on potential parameters and the number of CPU cores.

Main Results:

• ZMM performance is comparable or superior to SPME, despite using a larger cutoff distance.
• ZMM's efficiency stems from avoiding time-consuming electrostatic convolution and achieving shorter neighbor lists.
• Optimal ZMM performance is achieved with quadrupole/octupole cancellation and no damping factor.

Conclusions:

• ZMM is a highly efficient method for calculating electrostatic interactions in MD simulations.
• ZMM, particularly with specific configurations, presents a viable alternative for accelerating electrostatic calculations.
• The findings support ZMM as an excellent candidate for fast and accurate electrostatic computations in complex systems.