Free Energy Guided Sampling.

Area of Science:

• Computational Chemistry and Molecular Modeling
• Biophysics and Structural Biology

Background:

• Exploring the vast conformational space of molecules is crucial for understanding biological processes.
• Conventional molecular dynamics simulations often struggle with efficient sampling of rare events and complex energy landscapes.
• Existing methods may require predefined collective variables or reaction coordinates, limiting their applicability.

Purpose of the Study:

• To introduce and validate a new Free Energy-Guided Sampling (FEGS) method.
• To demonstrate FEGS's ability to accelerate conformational space exploration in unbiased molecular dynamics.
• To showcase FEGS's independence from collective variables or reaction coordinates.

Main Methods:

• Developed FEGS, a method utilizing cut-based free energy profiles and Markov state models.
• FEGS employs iterative parallel restarts from rarely visited free energy surface regions for exploration.
• A refinement stage initiates multiple runs from Boltzmann-distributed conformations.

Main Results:

• FEGS achieved 1 to 2 orders of magnitude speedup in conformational space exploration compared to conventional methods.
• Applied successfully to alanine dipeptide and a 20-residue β-sheet peptide, demonstrating broad applicability.
• The method preserved essential features of the free energy landscape, including basins and barriers.

Conclusions:

• FEGS significantly enhances the efficiency of molecular dynamics simulations for conformational sampling.
• The method's ability to control kinetic distance and its independence from reaction coordinates offer flexibility.
• FEGS provides a powerful tool for studying complex molecular systems governed by intricate energetic landscapes.