Analysis of Protein Dynamics Simulations by a Stochastic Point Process Approach

Area of Science:

• Biophysics
• Computational Biology
• Statistical Mechanics

Background:

• Molecular dynamics (MD) simulations now achieve millisecond timescales, enabling atomic-level study of complex protein dynamics.
• Intermittent conformational transitions and transient high-energy states are crucial biological phenomena accessible via MD.

Purpose of the Study:

• To develop a theoretical framework for efficiently extracting dynamical characteristics of point processes from MD trajectories.
• To enable accurate computation of the survival correlation function without assuming Poisson statistics.

Main Methods:

• Utilizing the stochastic theory of stationary point processes to analyze MD trajectories.
• Computing the survival correlation function from residence or interevent times for enhanced efficiency.
• Analyzing statistical and binning errors for experimental comparison.

Main Results:

• A novel theoretical framework is presented for analyzing protein dynamics from MD simulations.
• The method allows for orders of magnitude more efficient computation of the survival correlation function.
• Demonstrated application to a 1 ms MD trajectory of BPTI, analyzing water exchange kinetics and conformational fluctuations.

Conclusions:

• The developed framework efficiently extracts essential dynamical information from MD simulations.
• This approach facilitates a deeper understanding of rare conformational transitions and their biological roles.
• Accurate error analysis is crucial for comparing MD simulations with experimental data.