The complex portal--an encyclopaedia of macromolecular complexes

Abstract

The IntAct molecular interaction database has created a new, free, open-source, manually curated resource, the Complex Portal (www.ebi.ac.uk/intact/complex), through which protein complexes from major model organisms are being collated and made available for search, viewing and download. It has been built in close collaboration with other bioinformatics services and populated with data from ChEMBL, MatrixDB, PDBe, Reactome and UniProtKB. Each entry contains information about the participating molecules (including small molecules and nucleic acids), their stoichiometry, topology and structural assembly. Complexes are annotated with details about their function, properties and complex-specific Gene Ontology (GO) terms. Consistent nomenclature is used throughout the resource with systematic names, recommended names and a list of synonyms all provided. The use of the Evidence Code Ontology allows us to indicate for which entries direct experimental evidence is available or if the complex has been inferred based on homology or orthology. The data are searchable using standard identifiers, such as UniProt, ChEBI and GO IDs, protein, gene and complex names or synonyms. This reference resource will be maintained and grow to encompass an increasing number of organisms. Input from groups and individuals with specific areas of expertise is welcome.

Figure 1.

Screenshot of the results for the GO search term ‘GO:0003824’ (catalytic activity). Search results are displayed as 10 per page and filters for refining the search are shown in the left-hand side panel. The number of hits for each filter option is given in parenthesis.

Figure 2.

Screenshot of the details page of the human haemoglobin HbA complex (EBI-9008420). The Cross References table has been shortened to fit in the figure. Clicking on the hyperlinks will open the details directly in the external database.