IntAct

IntAct
Name	IntAct
Type	biological database
Scope	molecular interactions
Established	2006
Owner	EMBL-EBI
Country	United Kingdom
Access	open

IntAct

IntAct is an open-source molecular interaction database and curation platform that aggregates, annotates, and disseminates experimentally derived protein–protein, protein–small molecule, and complex interaction data. Developed and maintained within major European life science infrastructures, the resource connects with multiple community standards and repositories to provide high-quality, curated interaction datasets for researchers in molecular biology, structural biology, pharmacology, and systems biology.

Overview

IntAct collects experimentally supported interactions from peer-reviewed publications and direct submissions, integrating evidence such as affinity purification, yeast two-hybrid, crosslinking mass spectrometry, and structural data. It interoperates with major life science organizations and resources including European Molecular Biology Laboratory, European Bioinformatics Institute, UniProt, Protein Data Bank, Gene Ontology Consortium, and Reactome to enable cross-referenced annotation. Curated entries are annotated according to community standards maintained by consortia such as the HUPO Proteomics Standards Initiative and the International Molecular Exchange Consortium, ensuring compatibility with resources like BioGRID, STRING, and MINT. The platform supports data consumers ranging from bench scientists at institutions like Max Planck Society and Francis Crick Institute to bioinformatics groups at universities such as University of Cambridge and Massachusetts Institute of Technology.

Data Content and Curation

Content in IntAct derives from manual curation of literature, direct submissions from projects and consortia, and aggregation of interaction datasets from partner databases. Curation staff annotate experimental methods (e.g., co-immunoprecipitation, surface plasmon resonance, cryo-electron microscopy), interaction stoichiometry, interaction topology, and organismal source using controlled vocabularies from resources including PSI-MI, Gene Ontology Consortium, and NCBI Taxonomy. Curators operate according to provenance and quality frameworks aligned with policies from entities like European Research Council and standards bodies such as ISO committees relevant to bioinformatics. Cross-references to sequence databases such as UniProt, structural entries at Protein Data Bank, and pathway annotations from Reactome and KEGG enrich each curated record. The database distinguishes binary interactions from co-complex associations and captures evidence codes used by projects like ENCODE and Human Protein Atlas where applicable.

Database Structure and Access

IntAct implements a relational and graph-like internal model that maps interactors, interactions, experiments, and publications into discrete, queryable objects annotated with controlled identifiers. It exposes programmatic access via RESTful web services, downloadable PSI-MI XML and MITAB formats, and bulk FTP dumps compatible with workflows employed at institutions like European Bioinformatics Institute and computational platforms such as Galaxy Project and Cytoscape. The user interface supports advanced search, network visualization, and export functions that integrate with visualization tools developed by groups including Broad Institute and EMBL. Authentication and data submission pipelines accommodate contributors from academic centers like Wellcome Sanger Institute and industrial partners in pharmaceutical clusters such as Cambridge Biomedical Campus.

Integration and Interoperability

Interoperability is achieved through adherence to community ontologies and identifier schemes: sequence identifiers from UniProt, taxonomy identifiers from NCBI Taxonomy, ontology terms from Gene Ontology Consortium and PSI-MI, and literature identifiers from PubMed and CrossRef. The database participates in federated exchange via the International Molecular Exchange Consortium to synchronize records with MINT, DIP, BioGRID, and other interaction resources maintained by organizations like EMBO and Cold Spring Harbor Laboratory. Integration with structural resources such as Protein Data Bank and modeling efforts at AlphaFold enhances interpretation of interfaces, while links to chemical databases like ChEBI and PubChem support small-molecule interaction annotation. Data licensing and attribution conform to principles advanced by Creative Commons and research data initiatives led by European Open Science Cloud.

History and Development

Development began in the mid-2000s within European bioinformatics programs and consolidated under the umbrella of institutions including European Molecular Biology Laboratory and European Bioinformatics Institute. The platform evolved through collaborations with projects funded by agencies like Wellcome Trust and the European Commission, and through community-driven efforts organized by the HUPO Proteomics Standards Initiative and the International Molecular Exchange Consortium. Major milestones include adoption of PSI-MI formats, integration into the EBI service portfolio, and the transition to modern APIs and web frameworks influenced by software engineering practices at organizations such as Apache Software Foundation and GitHub. Ongoing development reflects contributions from academic groups at University of Oxford, University College London, and industry partners participating in data sharing consortia.

Applications and Impact

IntAct underpins hundreds of publications in areas spanning interactomics, network pharmacology, disease mechanism elucidation, and biomarker discovery, informing studies at centers like Harvard Medical School, Stanford University, and Institut Pasteur. Researchers leverage IntAct data to construct interactome maps, validate predicted interactions from tools developed at European Bioinformatics Institute and Broad Institute, and integrate interaction evidence into pathway models used by Reactome and KEGG. Pharmaceutical research groups at companies within clusters like Cambridge Biomedical Campus and Biocentury use curated interactions for target validation and hit triage. The resource also supports teaching and training initiatives at organizations such as EMBL-EBI Training and contributes to community benchmarks organized by Critical Assessment of Protein Structure Prediction and other consortium-led challenges.

Category:Biological databases