Predicting the folding pathway of engrailed homeodomain with a probabilistic roadmap enhanced reaction-path algorithm

Area of Science:

• Computational Biology
• Biophysics
• Protein Dynamics

Background:

• Predicting protein-folding pathways is crucial for understanding protein function.
• Traditional atomistic dynamic simulations are computationally intensive and time-consuming.

Purpose of the Study:

• To develop an efficient computational method for predicting protein-folding pathways.
• To validate the method's ability to identify on-pathway intermediates.

Main Methods:

• Introduced MaxFlux, a variational optimization approach to predict protein-folding pathways.
• Utilized the probabilistic roadmap method (PRM) to solve the MaxFlux global optimization problem.
• Employed CHARMM19 and the EEF1 implicit solvation model for protein-solution interactions.

Main Results:

• Successfully simulated the folding pathway of the engrailed homeodomain.
• Provided direct evidence confirming a previously reported intermediate state as a genuine on-pathway intermediate.
• Demonstrated moderate CPU power demand, indicating computational efficiency.

Conclusions:

• MaxFlux-PRM offers an efficient alternative to time-consuming dynamic simulations for predicting protein-folding pathways.
• The method accurately identifies on-pathway intermediates, advancing our understanding of protein folding dynamics.
• This approach holds promise for accelerating research in protein structure and function prediction.