APID

APID provides integrated collections of experimentally validated protein-protein physical interactions (PPIs) mapped to proteomes across over 1,100 organisms to support analysis of protein interaction networks.

Key Features:

• Comprehensive Data Integration: Integrates PPI data from primary databases BIND, BioGRID, DIP, HPRD, IntAct, and MINT and incorporates experimentally resolved 3D structures from the Protein Data Bank (PDB) when interactions involve more than two distinct proteins, totaling a curated collection of over 90,379 distinct proteins and 678,441 unique PPIs.
• Quality Control and Data Curation: Retains only experimentally validated PPIs, annotates interactions with their experimental validation methods to enable quality-level filtering, and refines PPI properties by eliminating false duplicates.
• Methodological Annotation of Interactions: Categorizes interaction detection methods into 'binary' (direct) and 'indirect' types and uses HUPO PSI-MI controlled vocabularies for standardized representation.
• Network Visualization and Sub-interactome Construction: Provides graph-based visualization capabilities based on Cytoscape.js and supports construction of sub-interactomes from query protein lists with annotation of interaction reliability and functional information.

Scientific Applications:

• Proteomics Research: Enables exploration of protein functions and interaction contexts within cellular proteomes using detailed PPI networks.
• Systems Biology: Supplies comprehensive interaction datasets for modeling biological systems and network-level analyses of cellular processes.
• Drug Discovery: Facilitates identification of candidate drug targets and interaction pathways relevant to disease mechanisms.
• Functional Genomics: Supports elucidation of gene function and regulatory mechanisms through functional annotations of proteins within interaction networks.

Methodology:

Unifies PPI data from BIND, BioGRID, DIP, HPRD, IntAct, and MINT; incorporates PDB-resolved 3D structures for interactions involving more than two distinct proteins; annotates experimental validation methods; categorizes detection methods into 'binary' and 'indirect' types; applies HUPO PSI-MI controlled vocabularies; and removes false duplicates while refining PPI properties.