From Human Salivary Proteome Wiki
Ontology Lookup
Many entities in the Human Salivary Proteome Wiki are annotated with terms from controlled vocabularies to describe their various properties. The use of these standard terms with precise definitions helps minimize ambiguity and facilitate information retrieval. Here we describe the Ontology Lookup function that you can use in the wiki to search, browse and read about the ontology terms.
See also: Help:Protein Annotation
Contents |
Introduction
Over the past decade or so, a large number of concepts have been carefully defined or standardized by biomedical subject matter experts. These sets of related concepts, known as controlled vocabularies or ontologies, provide very specific definitions of the biomedical terms to minimize potential ambiguity that the term itself may generate. Some of the ontologies also describe other properties of the concepts, group them into different semantic types (e.g. diseases, drugs, and genes), and show formal relationships that exist among them.
Benefits
There are multiple benefits of using standard terms for annotation of entities: searching for items with a specific property is more accurate because of the standard nomenclature of the annotations; there is a much lower likelihood of miss-understanding an annotation because well defined terms are used to capture the idea; it is easy to communicate to others what the annotation is as many other scientists use these same terms in their research; the concepts are frequently placed in a hierarchy of other concepts, showing how they are related to other terms (see below); and by placing these concepts into a hierarchy, computational reasoning can be performed on the annotations to create additional insights (e.g. if a protein is a member of certain protein family, it automatically inherits all the common functions that other family members have).
Browsing and Searching Ontologies
Your access point for searching and browsing ontologies contained in this wiki is the Ontology Lookup special page (Figure 1). You can go to this page by following the Ontology browser link under Ontologies in the navigation menu. Using this page you can either "Search" (for names and accession numbers) or "Browse" the ontology terms.
As you can see from Figure 1, there are a large set of ontologies currently available in the wiki. They are listed below along with links to their sites and the authors:
- BioCarta Pathway (BioCarta); BioCarta (local ontology)
- BRENDA Tissue/Enzyme Source (BTO); Technische Universität
- Cell Type (CL); OBO Foundry
- Chemical Entities of Biological Interest (CHEBI); OBO Foundry
- Gene Ontology (GO); Gene Ontology Consortium
- Human Disease (DOID); University of Maryland School of Medicine
- Human Phenotype Ontology (HP); OBO Foundry
- Kyoto Encyclopedia of Genes and Genomes (KEGG); Kyoto University and the University of Tokyo (local ontology)
- Mass Spectroscopy CV (PSI-MS) (PSI); OBO Foundry
- Medical Subject Headings (MSH); National Library of Medicine
- Pathway Ontology (PW); Medical College of Wisconsin
- UniProt Keyword list (UniProt); UniProt
- UniProt Taxonomy Database (NEWT); UniProt
Most of these ontologies are updated automatically to provide the most current listings; any change in the data (eg. additional new terms, updates to old terms) is immediately represented. Of the ontologies retrieved through Web services, most come from the EBI Ontology Lookup Service.
Browsing
You can browse all the ontologies individually to explore the type of information that is available. Figure 2 shows an example for Gene Ontology (GO). As you can see, GO starts with three major branches, "molecular function", "biological process" and "cellular component". This can be true of many ontologies, however some ontologies are more flat, only having one major level of detail (e.g. BioCarta).
You can press the "+" and "-" buttons in front of a term to respectively view or hide the descendants of the term. Clicking on the term itself and you will be taken to the detailed ontology term page for that term (see below).
Searching
You can search for terms by their names or accession numbers. Each approach will be described below:
By Name
If you don't know the specific location in a specific ontology that a term resides (very frequent), you can search for the term using the Name attribute in the Input Type field. Figure 3 shows the results of searching with the term "saliv" across all ontologies. If you want to search only a specific ontology, select that ontology from the drop-down box.
There are many things of note in Figure 3. First, the search finds any concepts that contain the search term, in this case "salivary" and "saliva" were both found (highlighted). Second, the concepts can come from many different ontologies, in front of each term is the ontology abbreviation for that ontology (see table above). Third, there are terms that don't contain "saliv" in their title. This is because synonyms of the concepts are also searched, as is the case with "Sialorrhea", which has a synonym of "Excessive salivation". Finally, between the "ontology abbreviation" and the ontology term is the "ontology accession number", which the unique identifier of this term in the ontology.
By Accession
Each ontology term has an accession number. This is a string of alphanumeric that is used to uniquely identify this term in the ontology of interest. To search for an ontology term by accession, you need to specify the ontology by selecting it from the Source drop-down box. You can then enter the accession number into the Search Term box. This must be an exact match for the accession number, partial matches will not work. Figure 4 shows the results searching for "12892" in the "Human Disease" ontology (the accession number for "Sialorrhea").
Ontology Term Details
If you want to see the specific details of a specific term, click on that term in the result list. Figure 5 shows the details of the ontology term "Cysteine Endopeptidases".
The page displays the source ontology, the accession number (a unique identifier) of the term, its preferred name ("Term Name"), the definition, any synonyms it may have, its relationships with other terms in the ontology, etc. For some ontologies, a link to the official page of the term is also provided.
The term relationships of a concept can be very important as it helps to determine whether there is a more appropriate term to describe the entity that you want to annotate. With this example, you can see that "Cysteine Endopeptidases" is a type of "Endopeptidases" (because it is it's "Parent"). There are also other types of endopeptidases, including "Cathepsins" and "Metalloendopeptidases" (because these are "Siblings" of cysteine endopeptidases). There are also more specific classes of cysteine endopeptidases, including "Bromelains" and "Caspases" (because these are "Children" of cysteine endopeptidases). It is from these relationships that you can learn a lot about how a term fits into a group of concepts. This also allows semantic reasoning to take place to discover new relationships between items that are described by these ontological terms.
References
Côté RG, et al. (2008) The Ontology Lookup Service: more data and better tools for controlled vocabulary queries. Nucleic Acids Res. 36(Web Server issue):W372-6 [PubMed:18467421]
