A divide-and-conquer strategy to improve diffusion sampling in generalized ensemble simulations.

Area of Science:

• Computational Chemistry
• Statistical Mechanics
• Molecular Dynamics

Background:

• Generalized ensemble simulations face diffusion-sampling problems, limiting efficiency with increasing system complexity.
• High entropic barriers significantly reduce sampling efficiency in complex conformational spaces.

Purpose of the Study:

• To develop a robust sampling strategy for generalized ensemble simulations.
• To enhance sampling efficiency by decomposing conformational space.

Main Methods:

• A novel divide-and-conquer sampling strategy based on simulated scaling.
• Decomposition of target conformational space into subconformational regions.
• Parallel simulations with Monte Carlo-based structure exchange (simulated scaling based variant Hamiltonian replica exchange method).

Main Results:

• The proposed method significantly improves sampling efficiency in generalized ensemble simulations.
• Demonstrated superior sampling capability compared to existing methods.
• Effectively addresses system size limitations in generalized ensemble simulations.

Conclusions:

• The simulated scaling based variant Hamiltonian replica exchange method offers a powerful solution to the diffusion-sampling problem.
• This approach can overcome critical bottlenecks in the development of generalized ensemble methods.
• Enables more efficient exploration of complex conformational landscapes in molecular simulations.