A theoretical framework for random acceleration molecular dynamics simulations

Area of Science:

• Computational chemistry
• Biophysics
• Molecular dynamics

Background:

• Ligand-receptor dissociation is a slow process, challenging standard simulation methods.
• Specialized techniques like Random Acceleration Molecular Dynamics (RAMD) accelerate these simulations.
• A theoretical framework for unbiased analysis of RAMD results is currently lacking.

Purpose of the Study:

• To develop a theoretical framework for interpreting RAMD simulations.
• To derive an analytical expression for unbiased escape times from RAMD data.
• To validate the theory using protein-ligand dissociation simulations.

Main Methods:

• Approximating RAMD simulations with overdamped Langevin dynamics.
• Utilizing a Smoluchowski equation with an effective temperature and diffusion coefficient.
• Deriving an analytical expression for unbiased escape time over a free-energy barrier.

Main Results:

• The RAMD model dynamics are consistent with a modified Smoluchowski equation.
• Effective temperature and diffusion coefficients depend quadratically on the random force magnitude.
• An analytical expression for unbiased escape time was derived and validated.

Conclusions:

• The developed theory provides a sound basis for interpreting RAMD simulations.
• The framework allows for the extraction of unbiased estimates from accelerated molecular dynamics.
• This approach has potential applications in analyzing protein-ligand complex dissociation.