The Arch from the Stones: Understanding Protein Folding Energy Landscapes via Bioinspired Collective Variables

Area of Science:

• Biophysics
• Computational Biology
• Molecular Dynamics

Background:

• Protein folding simulations face challenges due to timescale limitations.
• Effective collective variable (CV) design is crucial for enhanced sampling in simulations.
• Accurately capturing thermodynamics and intermediates requires overcoming these limitations.

Purpose of the Study:

• To introduce a strategy for automatically constructing complementary, bioinspired CVs.
• To enhance state resolution by capturing local hydrogen bonding and side-chain packing.
• To validate the CV strategy against unbiased simulations for protein folding.

Main Methods:

• Automatic construction of bioinspired collective variables (CVs).
• Distinguishing protein-protein from protein-water hydrogen bonding interactions.
• Incorporating native and non-native side-chain contacts.
• Utilizing advanced enhanced sampling methods with the designed CVs.

Main Results:

• Accurate resolution of complex free-energy landscapes for Chignolin and TRP-cage folding.
• Identification of critical folding intermediates, including the dry molten globule.
• Validation of the CV approach against extensive unbiased simulations.
• Demonstrated agreement with reference data.

Conclusions:

• The developed CV strategy is interpretable and portable.
• Microscopic details play a critical role in protein folding.
• This approach offers a promising avenue for studying larger and more complex biomolecular systems.