Contact-Map-Driven Exploration of Heterogeneous Protein-Folding Paths.

Area of Science:

• Computational Biology
• Biophysics
• Structural Biology

Background:

• Protein folding pathways are crucial for understanding protein function and dysfunction.
• Predicting protein folding mechanisms, especially those with multiple distinct pathways, remains a significant challenge.
• Previous methods relied on molecular dynamics (MD) simulations, which can be computationally intensive.

Purpose of the Study:

• To enhance a previously developed contact-map-based protein folding strategy.
• To accurately and robustly predict heterogeneous protein folding paths.
• To demonstrate the enhanced framework's ability to identify alternative folding mechanisms in a challenging multifolding-pathway protein.

Main Methods:

• Developed a novel topologically informed metric for comparing protein contact maps.
• Reformulated the graph-represented folding path generation process.
• Introduced a new, more reliable structural back-mapping algorithm for converting contact maps to Cartesian coordinates.
• Generated protein-folding trajectory ensembles without direct molecular dynamics simulations.

Main Results:

• The enhanced algorithm significantly improves the reliability of generating structurally sound folding intermediates.
• Physically irrelevant folding intermediates generated by the previous strategy were dramatically decreased.
• The enhanced method successfully identified alternative folding mechanisms for a multifolding-pathway protein.
• Results align with findings from direct molecular dynamics simulations.

Conclusions:

• The enhanced contact-map-based strategy provides an accurate and robust method for predicting complex protein folding pathways.
• This approach offers a computationally efficient alternative to molecular dynamics for studying protein folding mechanisms.
• The framework is capable of dissecting intricate folding landscapes, including those with heterogeneous secondary structural elements.