A new approach for extracting information from protein dynamics.

Area of Science:

• Biophysics
• Network Science
• Structural Biology

Background:

• Understanding protein dynamics is critical as protein structure prediction advances.
• Network science offers powerful tools for analyzing complex biological systems.
• Existing methods for protein dynamics networks often rely on limited correlation measures.

Approach:

• Applied an inverse approach to protein dihedral angle dynamics, bypassing the need for structural alignment.
• Utilized internal coordinates to represent protein dynamics.
• Developed a robust method for network extraction from molecular dynamics.

Key Points:

• The inverse approach generates physically interpretable and robust protein dynamics networks.
• Identified relevant allosteric pathways in the adhesion protein FimH.
• Detected subtle dynamical differences in structurally similar proteins like Siglec-8 and SARS-CoV-2 spike protein.

Conclusions:

• The inverse network approach provides significant biophysical insights into protein dynamics.
• This method enhances the analysis of allosteric mechanisms and functional differences.
• Offers a more robust alternative to correlation-based network analysis in structural biology.