Computing energy landscape maps and structural excursions of proteins

Area of Science:

• Computational Biology
• Biophysics
• Protein Dynamics

Background:

• Protein structural excursions are vital for biological recognition and function.
• Simulating protein dynamics is computationally intensive due to timescale disparities.
• Stochastic optimization offers structural insights but not dynamics.

Purpose of the Study:

• To develop a novel computational methodology for mapping protein energy landscapes.
• To identify energetically feasible structural excursions between protein states.
• To provide a practical approach for characterizing protein equilibrium dynamics.

Main Methods:

• Constructing multi-dimensional protein energy landscape maps using an evolutionary algorithm.
• Analyzing computed maps to identify key features like energy basins and barriers.
• Employing a path searching algorithm on a nearest-neighbor graph to find basin-to-basin excursions.

Main Results:

• The methodology successfully maps complex energy landscapes and identifies stable/semi-stable regions.
• Visual analysis reveals known and novel features of protein structure space.
• Comparison of sequence variants highlights the link between structure, dynamics, and function.

Conclusions:

• The proposed method efficiently locates basins and computes protein excursions within a practical computational budget.
• It generates testable hypotheses regarding mutation effects on protein function.
• This computational approach effectively guides experimental investigations in wet laboratories.