MiMiC: A Novel Framework for Multiscale Modeling in Computational Chemistry.

Area of Science:

• Computational Chemistry
• Multiscale Modeling
• Molecular Dynamics

Background:

• Developing efficient computational frameworks for multiscale modeling is crucial in computational chemistry.
• Existing methods often face limitations in flexibility and efficiency when coupling diverse simulation programs.
• Accurate treatment of long-range electrostatic interactions in hybrid QM/MM simulations remains a challenge.

Purpose of the Study:

• To present MiMiC, a novel and flexible framework for multiscale modeling.
• To enable efficient data exchange between loosely coupled programs for enhanced computational chemistry simulations.
• To facilitate advanced quantum mechanics/molecular mechanics (QM/MM) implementations for large systems.

Main Methods:

• Implementation of a multiple-program multiple-data (MPMD) model with loosely coupled programs.
• Utilization of MPI intercommunicators for fast data exchange between computational chemistry programs.
• Development of a new electrostatic embedding QM/MM approach coupling CPMD and GROMACS.

Main Results:

• MiMiC demonstrates high flexibility by allowing the coupling of various computational chemistry programs.
• The new QM/MM implementation efficiently treats long-range electrostatic interactions using QM multipoles.
• The framework supports partitioning systems into domains treated with different models (e.g., DFT, coarse-graining).

Conclusions:

• MiMiC provides a flexible and efficient platform for advanced multiscale modeling in computational chemistry.
• The developed QM/MM method significantly reduces computational cost for large system simulations without sacrificing accuracy.
• This framework enables large-scale molecular dynamics (MD) simulations, advancing the study of complex chemical systems.