GOAnnotator

GOAnnotator automates protein function annotation by retrieving and re-ranking relevant literature to identify Gene Ontology (GO) terms.

Key Features:

• PubRetriever: A hybrid system for retrieving and re-ranking relevant literature for protein function annotation.
• GORetriever+: An enhanced module for identifying Gene Ontology (GO) terms from retrieved texts.
• Automated literature retrieval: Retrieves relevant literature without relying on expert-curated or manually curated sources.
• Expanded protein coverage: Annotates proteins beyond those covered by existing text-based methods that depend on curated literature.

Scientific Applications:

• Automated protein function prediction: Predicts protein functions for understanding biological processes and advancing biomedical research.
• Gene Ontology term assignment: Assigns GO terms to proteins based on analysis of automatically retrieved scientific literature.
• Large-scale protein annotation: Streamlines annotation of protein functions across the UniProt database without manual curation.

Methodology:

GOAnnotator consists of two modules: PubRetriever, a hybrid system for retrieving and re-ranking relevant literature, and GORetriever+, an enhanced module for identifying Gene Ontology (GO) terms from the retrieved texts. Extensive experiments were conducted over three benchmark datasets.