PhyloCorrelate

PhyloCorrelate infers functional associations between bacterial genes by large-scale phylogenetic profiling to predict gene function and gene-gene interactions.

Key Features:

• Phylogenetic Profiling: PhyloCorrelate employs phylogenetic profiling to detect pairs of gene or protein families with similar evolutionary distributions across the bacterial tree of life, suggesting potential interactions, pathway memberships, or broader functional associations.
• Comprehensive Database: The tool utilizes a pre-computed database containing phylogenomic correlations for 27,372 gene families across 28,315 species and leverages Genome Taxonomy Database (GTDB) and AnnoTree data, incorporating entries from KEGG, PFAM, and TIGRFAM.
• Co-occurrence Metrics: PhyloCorrelate integrates multiple co-occurrence metrics, combining standard correlation measures and model-based approaches that account for phylogenetic history to produce an optimized scoring system linking genes with overlapping Gene Ontology (GO) terms and KEGG pathways.
• Extensive Comparisons: It conducts all-by-all comparisons of gene occurrence profiles across the bacterial tree of life, resulting in over 154 million comparisons for 28,315 genes across 27,372 bacterial genomes.
• High-Confidence Associations: The framework identified 29,762 high-confidence associations between bacterial gene/protein pairs and generated functional predictions for 834 domains of unknown function (DUFs) and proteins with unknown functions.

Scientific Applications:

• Gene function prediction: Predicting functions of bacterial genes and proteins, including proteins of unknown function and DUFs, based on correlated evolutionary patterns.
• Functional genomics and pathway inference: Inferring metabolic and signaling pathway memberships through co-occurrence links and shared KEGG pathway annotations.
• Interaction discovery: Identifying putative gene-gene interactions and functional associations across diverse bacterial lineages via correlated presence/absence profiles.
• Antimicrobial target prioritization: Prioritizing candidate targets for antimicrobial development using conserved co-occurrence and functional association signals.

Methodology:

Methods explicitly include phylogenetic profiling, a pre-computed phylogenomic correlation database, integration of multiple co-occurrence metrics (standard correlation measures and model-based approaches accounting for phylogenetic history), and all-by-all comparisons of gene occurrence profiles.