Sequence-structure matching in globular proteins: application to supersecondary and tertiary structure determination

Area of Science:

• Structural Bioinformatics
• Computational Biology
• Biochemistry

Background:

• The inverse protein folding problem, identifying sequences for a given 3D structure, remains a challenge.
• Existing methods struggle to identify proteins with similar structures but low sequence homology.

Purpose of the Study:

• To develop and validate a robust methodology for solving the inverse protein folding problem.
• To extend sequence-structure matching to identify proteins with divergent sequences but conserved structures.
• To enhance the prediction of local supersecondary structures like alpha/beta/alpha fragments and beta-hairpins.

Main Methods:

• Utilized a library of "protein fingerprints" based on side chain interaction patterns.
• Developed an exhaustive database search to match structures with sequences.
• Extended the methodology to accommodate insertions and deletions for identifying distantly related proteins.
• Applied sequence-structure comparison to predict supersecondary structures.

Main Results:

• Successfully matched protein structures to their corresponding sequences.
• Identified proteins with similar structures but low sequence homology, including plastocyanin/azurin, globins, proteases, cytochromes, actinidin/papain, and lysozyme/alpha-lactalbumin.
• Demonstrated high fidelity in predicting the location of alpha/beta/alpha fragments and beta-hairpins.

Conclusions:

• The developed methodology effectively addresses the inverse protein folding problem.
• This approach significantly enhances the prediction of global structural homology and local supersecondary structures.
• The method offers a powerful tool for protein sequence-structure relationship analysis and prediction.