COMPARTMENTS: unification and visualization of protein subcellular localization evidence

Area of Science:

• Proteomics
• Bioinformatics
• Cell Biology

Background:

• Protein subcellular localization is crucial for understanding cellular functions.
• Current methods for determining localization involve manual curation, microscopy, and sequence prediction, requiring consultation of multiple resources.
• A unified approach is needed for a comprehensive view of protein localization.

Purpose of the Study:

• To present the COMPARTMENTS resource, a unified database for protein subcellular localization information.
• To integrate diverse data sources including manual curation, high-throughput screens, sequence prediction, and automated text mining.
• To enhance data comparability and accessibility for researchers.

Main Methods:

• Integration of data from multiple protein localization databases and prediction tools.
• Implementation of automatic text mining for extracting localization information.
• Mapping all evidence to common protein identifiers and Gene Ontology (GO) terms.
• Assignment of confidence scores to localization evidence based on type and source.

Main Results:

• The COMPARTMENTS resource provides a comprehensive, integrated view of protein subcellular localization.
• All localization evidence is standardized using common identifiers and GO terms.
• Confidence scores facilitate the comparison of different data types and sources.
• A visualization tool offers a simple overview of unified localization evidence on a schematic cell.

Conclusions:

• COMPARTMENTS offers a valuable, up-to-date resource for protein subcellular localization.
• The integration and standardization of data improve the accessibility and utility of localization information.
• Confidence scoring and visualization aid in interpreting and comparing diverse localization evidence.