Rapid exploration of configuration space with diffusion-map-directed molecular dynamics.

Area of Science:

• Computational Chemistry
• Biophysics
• Molecular Modeling

Background:

• Molecular dynamics (MD) simulations are computationally limited in exploring the behavior of macromolecular systems.
• This limitation hinders understanding experimental results and predicting inaccessible phenomena.

Purpose of the Study:

• Introduce a novel sampling scheme, diffusion-map-directed MD (DM-d-MD), for accelerated exploration of molecular configuration space.
• Demonstrate DM-d-MD's ability to enhance computational efficiency in molecular simulations.

Main Methods:

• Developed DM-d-MD, a method that utilizes diffusion maps to guide MD simulations in real-time.
• Combined DM-d-MD with umbrella sampling to reconstruct equilibrium free energy landscapes.
• Tested the method on alanine dipeptide and alanine-12 systems.

Main Results:

• Achieved significant speedups in exploring configuration space compared to standard MD.
• Demonstrated a 3-order-of-magnitude speedup for alanine-12 at 300 K.
• Successfully reconstructed equilibrium distributions and free energy landscapes.

Conclusions:

• DM-d-MD offers a powerful, reaction coordinate-free approach for efficient molecular simulations.
• The method requires minimal prior knowledge, making it broadly applicable.
• DM-d-MD is expected to benefit studies of macromolecular systems where standard MD is computationally prohibitive.