There are three types of data in the PID:
Between 2006 and 2012, the NCI-Nature Curated data were created by Nature Publishing Group editors and reviewed by experts in the field. Biomolecules are annotated with UniProt protein identifiers and relevant post-translational modifications. Interactions are annotated with evidence codes and references.
BioCarta data from June 2004 was imported without expert review, and biomolecules are annotated by Entrez Gene identifiers without associated post-translational modifications. Evidence codes and references are not included in the annotation.
Human pathways from Reactome were first imported into the PID in December, 2007, and correspond to Reactome's Version 22 release. Biomolecules are annotated with UniProt identifiers and post-translational modifications.
All NCI-Nature Curated data are produced by Nature Publishing Group editors according to the following principles:
- Human model system: We focus on human data. Interactions in other mammals that are inferred to occur in humans are sometimes included with appropriate evidence codes.
- Biological relevance: Meaningful networks of undisputed interactions are synthesized into predefined pathways. Pathways selected for curation are based on suggestions made by our users, potential drugs targets and other biomolecules we know to be of interest to researchers.
- Authority: Molecular interactions are identified in primary peer-reviewed literature. Editors judge whether an interaction is physiologically relevant and assign evidence codes to each interaction. All pathways are reviewed by several experts in the field for accuracy and completeness.
- Consistent nomenclature: We use HUGO gene symbols, UniProt names or aliases, or Entrez Gene names or aliases for molecules. We use Gene Ontology controlled vocabulary terms for biomolecule and biological process annotation.