From residue coevolution to protein conformational ensembles and functional dynamics

Area of Science:

• Computational Biology
• Structural Biology
• Biophysics

Background:

• Protein structure prediction benefits from evolutionary amino acid correlations.
• Understanding protein function requires analyzing ensembles of structures, not just static ones.
• Enzymatic activity involves dynamic conformational changes like substrate binding and product release.

Purpose of the Study:

• To investigate protein conformational heterogeneity using coevolutionary data and molecular dynamics simulations.
• To develop a method for converting coevolutionary contact predictions into protein structural ensembles.
• To analyze the relationship between residue correlations and protein conformational diversity.

Main Methods:

• Adaptation of the Boltzmann-learning algorithm for homologous protein sequence analysis.
• Development of a coarse-grained protein model for contact prediction to structural ensemble conversion.
• Exhaustive sampling simulations to analyze conformational ensembles consistent with residue correlations.

Main Results:

• The most representative protein structure aligns with experimental folds.
• Different sequence regions exhibit varying stability, linked to biologically relevant conformations.
• Coevolving pairs show cooperativity influencing conformational stability and transitions.

Conclusions:

• The developed protocol accurately reproduces key aspects of protein folding mechanisms.
• It effectively samples conformers involved in protein conformational transitions.
• This approach enhances the understanding of protein dynamics and functional states.