Cellular Component Ontology Guidelines

The cellular component ontology describes locations, at the levels of subcellular structures and macromolecular complexes. Examples of cellular components include 'nuclear inner membrane', with the synonym 'inner envelope', and the 'ubiquitin ligase complex', with several subtypes of these complexes represented.

Generally, a gene product is located in or is a subcomponent of a particular cellular component. The cellular component ontology includes multi-subunit enzymes and other protein complexes, but not individual proteins or nucleic acids. Cellular component also does not include multicellular anatomical terms.

Integration with SAO (Subcellular Anatomy Ontology)

The primary use of the GO Cellular Component Ontology is for GO annotation, but it has also been used for phenotype annotation, and for the annotation of images. Another ontology with similar scope is the Subcellular Anatomy Ontology (SAO), part of the Neuroscience Information Framework Standard (NIFSTD) suite of ontologies. The SAO also covers cell components, but in the domain of neuroscience.

Recently, the GO Cellular Component Ontology was enriched in content and links to the Biological Process and Molecular Function branches of GO as well as to other ontologies. This was achieved in several ways, one of which was amalgamation of SAO terms with GO Cellular Component ones. As a result, nearly 100 new neuroscience-related terms were added to the GO.

A recent paper describes this effort, along with other recent developments in the GO Cellular Component Ontology.

Maintaining complete 'is_a' and 'part_of' trees in cellular component

The cellular component ontology is is_a complete, meaning that every term has a path to the root node which passes solely through is_a relationships. This should be preserved; the following guidelines should help maintain this structure.

Make sure the term has an is_a path to the root, i.e. there are is_a parent terms by at least one path all the way to 'cellular component'. Make sure the term has at least one part_of relation in its ancestry, to ensure that there are no part_of orphans. It does not need to be an immediate part_of parent, but every term has to be part_of something. So, for complex Y, this would be okay for example:

  • cell
    • [p] complex X
      • [i] complex Y
because complex Y is transitively part_of cell. Ensure that all logical is_a parents are added. So, for example, if your term is a protein complex, make sure it has the parent 'protein complex'. Or if your term, or one of its parents, is part_of cell, it will need to be is_a 'cell part', or have 'cell part' in its ancestry. Check none of the relations you create are redundant. You can check for this in OBO-Edit by using the reasoner, and then using the link filter [self] [self] [is redundant]. As an added check, a weekly job runs every Monday night (US West Coast time) to remove any redundant relationship.