The network of stabilizing contacts in proteins studied by coevolutionary data

Area of Science:

• Computational biology
• Biophysics
• Structural biology

Background:

• Protein primary structure (amino acid sequence) is abundant experimental data.
• Co-evolutionary analysis of protein families can reveal native conformation.
• Inverse Ising models are used to study correlations in biomolecules.

Purpose of the Study:

• To optimize an algorithm for calculating effective residue interaction energies.
• To validate the optimized algorithm using model and real biological systems.
• To analyze the interaction network stabilizing protein native conformations.

Main Methods:

• Inverse Ising model approach.
• Optimization of algorithms for calculating effective energies.
• Back-calculation of interaction energies in model systems.
• Prediction of free energies for mutations in real systems.

Main Results:

• The optimized algorithm accurately calculates effective residue interaction energies.
• The approach successfully predicts mutation-associated free energies.
• The study identified distinct properties of the interaction network stabilizing protein native structures compared to contact networks.

Conclusions:

• The optimized inverse Ising model provides a robust method for inferring protein interaction networks.
• The identified interaction networks offer insights into protein stability and folding.
• This approach advances our understanding of structure-function relationships in proteins.