Transition Path Sampling Guided by Structural Motifs.

Area of Science:

• Computational biology
• Structural biology
• Biophysics

Background:

• Understanding dynamic protein function requires knowledge of conformational states and transitions.
• Determining all key conformational states for a single protein remains challenging.
• Leveraging conformational data from homologous proteins is difficult with low sequence identity.

Purpose of the Study:

• To develop a novel simulation technique for predicting protein conformational changes.
• To guide protein dynamics towards a target state using a structural similarity metric (SSM).
• To enable simulation of conformational transitions even with low sequence similarity to homologous proteins.

Main Methods:

• Developed a simulation technique using distance matrices to a target state to define an SSM.
• Applied SSM to guide protein folding from a known conformation to a target state.
• Generalized SSM to multiple dimensions based on structural motifs for full protein folding simulation.

Main Results:

• Successfully folded the β-hairpin portion of the β-β-α (BBA) protein to its native state using SSM.
• Demonstrated simulation of protein transitions to new conformations based on homologous structures with low sequence similarity.
• Compared different enhanced sampling methods coupled to SSM, discussing their advantages and disadvantages.

Conclusions:

• The SSM-based simulation technique effectively guides protein dynamics to target conformations.
• This method allows prediction of conformational changes in proteins with low sequence similarity to known structures.
• Deconstructing protein state space into structural motifs offers an efficient framework for exploring conformational landscapes.