NOA

NOA extends Gene Ontology annotations from genes to interactions (edges) within biological networks to perform network-based enrichment analysis.

Key Features:

• Link Ontologies: Creates link ontologies that assign Gene Ontology functions to network interactions using known annotations of the genes forming each edge.
• Coverage and Diversity optimization: Optimizes the 'Coverage' and 'Diversity' indices during assignment to ensure comprehensive and specific functional annotations.
• Alternative reference sets: Generates alternative reference sets to serve as background models for enrichment calculations.
• Statistical ranking: Statistically ranks enriched functional terms within networks using the generated reference sets.
• Network-level enrichment: Extends enrichment analysis from individual genes to edges to capture interaction-based functional signals.
• Network types: Applicable to transcription regulatory networks and protein interaction networks and able to detect dynamic functional changes across network types.
• Outputs: Produces result tables, heatmaps, and overview networks representing enrichment and functional annotations.
• Implementation: Implemented as an open-source Java plugin compatible with Cytoscape version 2.8 for annotating one or more biological networks based on user-defined parameters.

Scientific Applications:

• Functional annotation of interactions: Annotates molecular interactions with Gene Ontology terms to reveal edge-level functions.
• Network-based enrichment analysis: Identifies enriched GO terms at the network and edge level using alternative reference sets and statistical ranking.
• Comparative and dynamic analysis: Captures dynamic changes in functional signals across transcription regulatory and protein interaction networks.
• Improved specificity in static networks: Detects more relevant and specific functions in static networks compared to conventional gene-centric enrichment methods.

Methodology:

Creates link ontologies by assigning GO terms to interactions from joint-gene annotations, optimizes 'Coverage' and 'Diversity' indices for assignments, generates alternative reference sets, and statistically ranks enriched functional terms.