Combining the biased and unbiased sampling strategy into one convenient free energy calculation method

Area of Science:

• Computational Chemistry
• Molecular Dynamics
• Biophysics

Background:

• Calculating free energy landscapes for large molecules is computationally challenging.
• Existing methods often require high temperatures or strong driving forces to enhance sampling over free energy barriers.
• Efficiently exploring complex molecular systems remains a key objective in computational science.

Purpose of the Study:

• To develop a more convenient and efficient method for constructing free energy landscapes.
• To combine adaptive biasing and multiple temperature replicas into a single simulation framework.
• To simplify the process of free energy calculation without compromising sampling speed or data simplicity.

Main Methods:

• A mixed simulation method employing adaptive biasing potential on some molecular replicas.
• Utilizing unbiased and exchangeable replicas at various temperatures for canonical ensemble sampling.
• Implementing Monte Carlo trial moves for state variable transfer from biased to unbiased replicas post-equilibrium.

Main Results:

• The proposed method integrates acceleration strategies for enhanced sampling efficiency.
• It eliminates the need for an initial reference biasing potential or numerous replicas.
• Achieves sampling speeds comparable to biased simulations with the processing simplicity of unbiased simulations.

Conclusions:

• The developed mixed method offers a minimalist and user-friendly approach to free energy landscape construction.
• It significantly improves sampling efficiency and simplifies data processing for large molecular systems.
• Represents a practical advancement in computational techniques for molecular modeling and analysis.