PathFams: statistical detection of pathogen-associated protein domains

Area of Science:

• Microbiology
• Bioinformatics
• Genomics

Background:

• Many bacterial genes encode proteins with unknown functions.
• Identifying bacterial virulence factors is crucial for understanding pathogenesis.
• Locating virulence determinants within these proteins presents a significant challenge.

Purpose of the Study:

• To systematically identify protein domains associated with bacterial pathogenicity.
• To develop a resource for discovering novel virulence factors.
• To aid in the functional annotation of bacterial genomes.

Main Methods:

• Comprehensive analysis of 17,929 protein domain families from the Pfam database.
• Scoring domains based on overrepresentation in pathogenic species, taxonomic distribution, and metagenomic abundance.
• Utilizing comparative genomics and bioinformatics approaches.

Main Results:

• Identification of specific pathogen-associated domain families.
• Discovery of candidate virulence factors in the human gut microbiome.
• Characterization of eukaryotic-like mimicry domains with potential roles in virulence.

Conclusions:

• The study successfully identified key protein domains linked to bacterial virulence.
• An interactive database, PathFams, was developed to facilitate exploration of these domains.
• PathFams enables users to identify pathogen-associated domains and architectures in custom sequences, supporting ongoing virulence factor research.