Estimation of protein folding probability from equilibrium simulations.

Area of Science:

• Computational biology
• Biophysics
• Molecular dynamics

Background:

• Understanding protein folding is crucial for molecular biology and drug discovery.
• Existing methods for calculating folding probabilities often require extensive simulations.
• Analyzing folding pathways provides insights into protein structure and function.

Purpose of the Study:

• To develop a computational procedure for evaluating folding probabilities (p(fold)) from existing molecular dynamics trajectories.
• To demonstrate the utility of this method in identifying multiple folding pathways.
• To apply the procedure to a specific peptide system.

Main Methods:

• Utilizing the assumption that similar structures share similar folding probabilities.
• Applying a structurally homogeneous clustering technique to molecular dynamics snapshots.
• Analyzing equilibrium folding-unfolding trajectories without additional simulations.

Main Results:

• Successfully evaluated p(fold) for each snapshot in a molecular dynamics trajectory.
• Demonstrated the detection of multiple folding pathways for a three-stranded antiparallel beta-sheet peptide.
• The method requires no additional simulations, making it computationally efficient.

Conclusions:

• The developed procedure provides an efficient way to assess protein folding probabilities and pathways.
• This method can reveal complex folding mechanisms previously difficult to detect.
• It offers a valuable tool for researchers studying protein dynamics and structure-function relationships.