Constructing the equilibrium ensemble of folding pathways from short off-equilibrium simulations

Area of Science:

• Computational Biology
• Protein Dynamics
• Biophysics

Background:

• Characterizing protein folding pathways and their probabilities is a key challenge in protein folding theory.
• All-atom molecular dynamics simulations can provide this information but are limited by the rare event nature of protein folding, requiring impractically long simulation times.
• Simplified protein models are often used to overcome simulation length limitations, but they may not accurately reflect real-world folding processes.

Purpose of the Study:

• To present a novel computational approach for reconstructing the complete ensemble of protein folding pathways.
• To enable the analysis of folding pathways from simulations significantly shorter than the actual folding time.
• To apply this method to all-atom simulations in explicit solvent without relying on predefined reaction coordinates.

Main Methods:

• Developed a method to reconstruct the full ensemble of folding pathways from short simulations.
• Partitioned the conformational state space into small states and constructed a Markov model between them.
• Applied the method to all-atom simulations of a PinWW domain in explicit solvent.

Main Results:

• Successfully reconstructed the full ensemble of folding pathways from simulations much shorter than the protein's folding time.
• Results for the PinWW domain folding showed good agreement with experimental kinetic data.
• Identified a complex and parallel folding process, including misfolded trap states that significantly impede folding speed.

Conclusions:

• The developed approach allows for efficient characterization of protein folding pathways using all-atom simulations in explicit solvent.
• The study provides detailed insights into the complex nature of protein folding, highlighting the impact of misfolded states.
• This method overcomes the limitations of simulation length for studying rare folding events.