Docking-based long timescale simulation of cell-size protein systems at atomic resolution

Area of Science:

• Computational Biology
• Molecular Modeling
• Biophysics

Background:

• Modeling whole cells at the molecular level is a growing area in computational biology.
• Protein interactions are central to cellular functions, but current modeling techniques have limitations.
• Existing methods like protein docking are fast but lack kinetic information, while molecular simulations are slow or coarse-grained.

Purpose of the Study:

• To develop a novel computational approach that bridges protein docking and molecular simulation.
• To achieve unprecedented simulation timescales at all-atom resolution for large protein systems.
• To accurately model protein interactions, including their dynamics and kinetics.

Main Methods:

• Developed a hybrid approach combining pairwise fast Fourier transform (FFT) docking with Monte Carlo simulations.
• Mapped the global intermolecular energy landscape of protein systems.
• Sampled the energy landscape in both space and time using Monte Carlo methods.
• Parametrized and validated the simulation protocol using experimental data and molecular dynamics simulations.

Main Results:

• The novel simulation protocol achieved second-long trajectories for cellular-scale protein systems at atomic resolution.
• The method demonstrated consistent performance across diverse protein systems and concentrations.
• Successfully recapitulated experimental and molecular dynamics data on protein diffusion rates and aggregation.

Conclusions:

• This proof-of-concept study presents a significant advancement in modeling protein interactions.
• The developed approach overcomes limitations of existing methods, offering speed and atomic detail.
• Enables large-scale, long-timescale simulations of protein systems, crucial for understanding cellular processes.