Functional fingerprints of folds: evidence for correlated structure-function evolution

Area of Science:

• Structural bioinformatics
• Evolutionary biology
• Computational biology

Background:

• Protein domains are fundamental units of proteins, possessing distinct structures and functions.
• Understanding the evolutionary interplay between protein structure and function is crucial for deciphering protein evolution.
• Existing methods often analyze structure and function separately, limiting insights into their co-evolution.

Purpose of the Study:

• To investigate the relationship between protein domain structure and function using evolutionary perspectives.
• To determine if distinct structural clusters of protein domains possess unique functional profiles.
• To explore how evolutionary divergence impacts the structure-function relationship in protein domains.

Main Methods:

• Utilized the protein domain universe graph (PDUG) for structural similarity clustering of protein domains.
• Employed hierarchical functional annotation from Gene Ontology (GO) to analyze domain functions.
• Analyzed the distribution of functions within each structural cluster to identify functional fingerprints.

Main Results:

• Each structural cluster (domain fold) exhibits a unique distribution of functions, termed a 'functional fingerprint'.
• These functional fingerprints are specific to structural clusters and vary significantly between them.
• Relaxing the structural similarity threshold in PDUG allowed earlier-diverged domains to join clusters without compromising their functional fingerprints.

Conclusions:

• Protein domain structure and function co-evolve, with unique functional fingerprints characterizing distinct structural folds.
• The observed patterns support a divergent evolution model where structural and functional traits evolve in concert from common ancestors.
• This study provides a framework for understanding protein evolution through the lens of integrated structural and functional analysis.