PANNZER

PANNZER annotates protein sequences with Gene Ontology (GO) terms and free-text functional descriptions to improve functional annotation of proteins in databases such as UniProtKB.

Key Features:

• Functional annotation prediction: Predicts Gene Ontology (GO) classes and free-text functional descriptions for protein sequences.
• High-throughput capability: Designed to perform large-scale annotation of numerous protein sequences.
• Advanced methodology: Employs a weighted k-nearest neighbour (k-NN) approach combined with statistical testing.
• Benchmark performance: Demonstrated superior performance for free-text description prediction in CAFA 2011 and improved GO prediction over previous iterations.

Scientific Applications:

• Functional genomics: Annotates uncharacterized proteins in newly sequenced genomes to support genome-scale functional characterization.
• Protein engineering and drug discovery: Provides predicted functions and descriptions to inform target selection and hypothesis generation.
• Systems biology: Improves the functional annotation of proteins within biological networks and pathways.

Methodology:

Weighted k-nearest neighbour (k-NN) algorithm combined with statistical testing.