Defining structural and evolutionary modules in proteins: a community detection approach to explore sub-domain...

Area of Science:

• Structural Biology
• Computational Biology
• Protein Science

Background:

• Protein modularity is crucial for understanding protein evolution and function.
• Identifying sub-domain architecture within proteins remains a challenge.
• Existing methods often focus on sequence motifs rather than structural correlations.

Purpose of the Study:

• To develop and validate a robust graph-theory approach for identifying protein sub-domain architecture.
• To statistically assess the significance of identified modules.
• To analyze both evolutionary and dynamic modularity in protein structures.

Main Methods:

• Utilized graph theory to optimize partitions maximizing modularity score.
• Applied significance and power testing to identify statistically supported modules.
• Analyzed homologous structures for evolutionary modules and molecular simulations for dynamic modules.

Main Results:

• Successfully identified modules in simulated data, even with low internal landmark correlation.
• Discovered four robust evolutionary modules in alpha-amylase, including the minimal functional TIM barrel.
• Inferred dynamic modules in NPC1 protein that correlate with functional components and disease relevance.

Conclusions:

• Demonstrated a novel method to robustly identify statistically supported sub-domain architecture.
• The approach reveals functional/structural insights, exemplified by alpha-amylase and NPC1 proteins.
• This method offers a new perspective on protein compartmentalization through residue correlation analysis.