• Home
• Search
• Downloads
• Help
• Tutorial
• Contact Us

• Help
• ComPPI for Network Scientists

ComPPI is a multicomponent, undirected, weighted protein-protein interaction network.

• Nodes: proteins
• Edges: undirected edges representing any relationship between two proteins
• Edge weights: interaction scores

Node attributes

• Name: the protein name in the strongest available naming convention, usually UniProt/Swiss-Prot
• Naming convention: the naming convention of the protein name
• Synonyms: other names of the protein
• Taxonomy ID: the Gene Ontology Taxonomy ID of the related species, such as '9606' for human
• Localization Scores: the probability that the protein belongs to a given major subcellular localization
• Major Subcellular Localizations: cytosol, nucleus, etc.
• Minor Subcellular Localizations: the Gene Ontology cellular component IDs
• Localization Source Databases: the names of the source databases of the subcellular localization data
• Localization Source PubMed IDs: the PubMed IDs of the publications describing the source databases or the specific publication which relates the protein to a given subcellular localization
• Localization Experimental Systems: the experimental system that associated the protein with a given localization (such as 'annotated protein expression')
• Localization Experimental System Type: unknown (0), experimental (1), predicted (2)

Edge attributes

• Interaction Score: the probability of the given interaction, can be thought of as edge weight
• Interaction Source Databases: the names of the databases that describe the interaction
• Interaction Source PubMed IDs: the PubMed IDs of the publications describing the source databases or the specific publication which describes the interaction
• Interaction Experimental Systems: the experimental system that determined the interaction (such as 'yeast two-hybrid')

Notes

The custom downloads contain various combinations of these, please refer to the relevant help section for further details.

The nodes are identified by ComPPI IDs in the downloadable full database (SQL). These IDs are not stable, they may change in every release. If you use ComPPI to build a network, you may want to consider using the protein names as node IDs, otherwise networks of different ComPPI versions may not be compatible with each other.

⇦ Back to Help Table of Contents
⇦ Back to Starting Page