ART-RRT: As-Rigid-As-Possible search for protein conformational transition paths

Area of Science:

• Biophysics
• Computational Biology
• Structural Biology

Background:

• Protein function is intrinsically linked to conformational changes.
• Predicting these transitions is crucial for understanding biochemical mechanisms.
• Existing methods often yield physically unrealistic, self-intersecting paths for complex motions.

Purpose of the Study:

• To extend the ART-RRT method for predicting protein conformational transition pathways.
• To develop a computationally efficient approach for generating biologically relevant paths.
• To overcome limitations of existing geometry-based methods regarding self-intersection.

Main Methods:

• Adaptation of the ART-RRT algorithm, originally for ligand-unbinding pathways.
• Bidirectional exploration strategy rooted from the two stable end states.
• Introduction of a novel connection strategy between explored regions.
• Leveraging computer graphics-inspired mechanisms for structural consistency.

Main Results:

• Generation of probable conformational transition paths between protein stable states.
• Paths are computationally inexpensive to produce.
• Resulting pathways are biologically realistic and free from self-intersection.
• The method effectively explores energy valleys in low-dimensional spaces.

Conclusions:

• The adapted ART-RRT method provides a robust and efficient tool for studying protein conformational dynamics.
• Generated pathways can serve as valuable input for advanced protein analysis.
• This approach enhances the understanding of protein mechanisms by providing realistic transition path data.