Reduction of All-Atom Protein Folding Dynamics to One-Dimensional Diffusion

Area of Science:

• Computational Biology
• Biophysics
• Protein Dynamics

Background:

• Protein folding is crucial for biological function.
• Previous models simplified folding dynamics to diffusion on low-dimensional free energy surfaces.
• The accuracy and dimensionality requirements of these models were unclear.

Purpose of the Study:

• To determine if protein folding dynamics can be accurately described by diffusion on a single coordinate.
• To assess the accuracy of the fraction of native contacts (Q) as a folding coordinate.
• To investigate the possibility of improving folding models with optimized coordinates.

Main Methods:

• All-atom molecular dynamics simulations of ten small proteins in explicit solvent.
• Analysis of folding dynamics using the fraction of native contacts (Q) as a reaction coordinate.
• Development and testing of an optimized coordinate (Qopt) for improved model accuracy.

Main Results:

• Protein folding dynamics were accurately described by one-dimensional diffusion on the Q coordinate for most proteins studied.
• Diffusion models successfully reproduced folding rates, transition-path durations, and diffusive propagators.
• Diffusion coefficients decreased with increasing protein chain length, consistent with energy landscape theory.
• An optimized coordinate (Qopt) improved the accuracy of the diffusion model for proteins where the Q coordinate was insufficient.

Conclusions:

• One-dimensional diffusion on a suitable coordinate is a remarkably faithful model for the folding dynamics of small proteins.
• The fraction of native contacts (Q) serves as an effective, general coordinate for modeling protein folding.
• Optimized coordinates can further enhance the accuracy of diffusion-based folding models.