Systematic coarse-graining of molecular models by the Newton inversion method.

Area of Science:

• Computational Chemistry
• Molecular Modeling
• Materials Science

Background:

• Developing accurate coarse-grained (CG) models is crucial for simulating large-scale molecular systems.
• Existing methods face challenges in parameterization and preserving specific molecular properties.

Purpose of the Study:

• To systematically construct CG molecular models from detailed atomistic and ab initio simulations.
• To develop robust methods for parameterizing CG models using structural and thermodynamic data.
• To demonstrate the application of these CG models in simulating complex self-assembly phenomena.

Main Methods:

• Utilizing atomistic simulations to extract structural information for CG model parameterization.
• Inverting statistical-mechanical equations (e.g., Inverse Monte Carlo) to derive effective potentials.
• Developing novel approaches to address specific challenges like high pressure in water models.
• Creating enantiomerically specific CG models for molecules like prolines.
• Revisiting and updating existing CG models, such as lipid models, with improved force fields.

Main Results:

• Successfully generated CG models for water, prolines, and lipids with preserved properties.
• Developed an effective method to overcome the high-pressure problem in united atom water models.
• Created enantiomerically specific CG models for L- and D-prolines in dimethylsulfoxide.
• Demonstrated spontaneous formation of diverse structures (vesicles, micelles, multi-lamellar) using an updated CG lipid model in large-scale simulations.

Conclusions:

• Systematic construction of CG models from atomistic data is feasible and effective.
• The developed methods allow for accurate representation of molecular systems at the mesoscale.
• CG models are powerful tools for studying self-assembly and complex phase behavior under various thermodynamic conditions.