Mesoscopic-microscopic spatial stochastic simulation with automatic system partitioning

Area of Science:

• Computational chemistry
• Biophysics
• Chemical kinetics

Background:

• Reaction-diffusion master equation (RDME) enables efficient on-lattice simulation of spatially resolved stochastic chemical kinetics.
• RDME is faster than off-lattice methods for coarse lattices but struggles with diffusion-controlled reactions requiring fine meshes.
• Existing mesoscopic-microscopic hybrid methods require manual system partitioning, limiting their application.

Purpose of the Study:

• To develop a novel hybrid simulation algorithm for mesoscopic systems with multiscale properties.
• To enable automatic system partitioning for improved efficiency and accuracy in stochastic simulations.
• To overcome limitations of manual partitioning in current mesoscopic-microscopic hybrid methods.

Main Methods:

• Proposed a hybrid simulation algorithm integrating mesoscopic and microscopic scales.
• Implemented automatic system partitioning based on indirect a priori error estimates.
• Validated the method on 3D models of diffusion-controlled reaction networks.

Main Results:

• The developed hybrid method demonstrates accuracy and efficiency for complex systems.
• Automatic partitioning successfully addresses the multiscale nature of diffusion-controlled reactions.
• The algorithm provides a more 'black-box' approach compared to previous hybrid methods.

Conclusions:

• The proposed automatic partitioning hybrid method enhances simulation efficiency and accuracy for spatially resolved stochastic kinetics.
• This approach is particularly beneficial for systems with varying degrees of diffusion control.
• The method offers a significant advancement for simulating complex chemical kinetics in biological and chemical systems.