Simple Method for Simulating the Mixture of Atomistic and Coarse-Grained Molecular Systems

Area of Science:

• Computational chemistry
• Molecular dynamics simulations
• Biophysics

Background:

• Hybrid simulations combining fine-grained (FG) and coarse-grained (CG) models offer a way to overcome limitations of each approach.
• Existing methods often require extensive parametrization and lack transferability.
• Developing robust and transferable hybrid simulation protocols is crucial for studying complex biological systems.

Purpose of the Study:

• To develop and validate a simple protocol for combining FG and CG systems in molecular simulations.
• To investigate the interactions between FG protein atoms and CG water using virtual sites.
• To optimize the FG-CG mixing scheme for accurate protein structure and dynamics representation.

Main Methods:

• Utilized a hybrid simulation approach integrating standard FG and CG force field models.
• Employed virtual sites to bridge interactions between FG protein atoms and CG water.
• Tested the protocol on small protein systems, systematically adjusting parameters like Lennard-Jones potentials.
• Incorporated a position-restrained FG water layer to enhance structural stability.

Main Results:

• The conventional CG model (MARTINI) led to protein unfolding due to strong coupling with the FG model.
• Reducing Lennard-Jones interactions between CG atoms and virtual sites improved protein secondary and tertiary structure stability.
• Adding a thin layer of position-restrained FG water further enhanced structural mimicry of FG-only simulations.
• Free energy landscapes indicated overstabilization of the protein native structure, potentially limiting the study of protein dynamics.

Conclusions:

• A simple hybrid FG-CG simulation protocol using virtual sites and a restrained FG water layer can effectively stabilize protein structures.
• Parameter optimization, particularly for non-bonded interactions and solvent representation, is critical for accurate hybrid simulations.
• While structural stability is improved, potential overstabilization of the native state requires further investigation for applications involving protein dynamics.