EFI

EFI supports functional annotation of uncharacterized proteins by generating Sequence Similarity Networks (SSNs) with EFI-EST, constructing Genome Neighborhood Networks and Diagrams (GNNs/GNDs) with EFI-GNT, and performing Chemically Guided Functional Profiling (CGFP) with EFI-CGFP to prioritize enzyme function assignments.

Key Features:

• Sequence Similarity Networks (SSNs) / EFI-EST: Generates SSNs for protein families to explore relationships between proteins based on sequence similarity.
• Genome Neighborhood Networks and Diagrams (GNNs/GNDs) / EFI-GNT: Analyzes genomic context of proteins clustered in SSNs to infer potential functions from gene proximity and co-occurrence.
• Chemically Guided Functional Profiling (CGFP) / EFI-CGFP: Integrates metagenome abundance data to prioritize uncharacterized protein clusters for functional assignment.
• Databases used: Leverages protein, genome, and metagenome databases to support sequence- and structure-based functional annotation.
• Sequence- and structure-based strategy: Applies sequence- and structure-based analyses to support enzyme function predictions.
• Visualization outputs: Produces outputs compatible with Cytoscape and generates genome neighborhood diagrams (GNDs) and CGFP heatmaps summarizing metagenome abundance.
• Integrated workflow: Enables transfer and combined analysis of data across SSN, GNT, and CGFP stages.

Scientific Applications:

• Functional annotation: Assigns functions to uncharacterized proteins and unknown enzymes identified in genome projects.
• Operon and pathway inference: Predicts functional roles of genes based on genome neighborhood relationships and co-occurrence patterns.
• Candidate prioritization for experiments: Identifies high-priority protein clusters for experimental characterization using metagenome abundance as a guide.
• Metabolic and enzymatic pathway discovery: Supports discovery and mapping of enzymatic and metabolic pathways across genomes and metagenomes.
• Protein family analysis: Explores subfamilies and functional diversification within protein families using SSNs.

Methodology:

Computational steps include generating Sequence Similarity Networks with EFI-EST, constructing Genome Neighborhood Networks and Diagrams with EFI-GNT, and integrating metagenome abundance in EFI-CGFP, using protein, genome, and metagenome databases and sequence- and structure-based analyses.