Survey miniwebsite

We present the results of a survey carried out to establish which naming conventions are currently employed by
ontologies and to determine what their special requirements regarding the naming of entities might be.



The survey was conducted by contacting the custodians of the 66 OBO ontologies (as of November 2007) either by email or telephone. Each respondent then received a questionnaire, that
can be downloaded here:

The overall response rate has been excellent and the Table 1 lists the ontologies that have participated in this survey. Of the 66 OBO ontologies contacted, 42 questionnaires were returned. The apparent failure of some groups to respond to the questionnaire is due to one of the following reasons, documented in the \'Comment\' column; (i) two or more artifacts belong to the same ontology (i.e., GO exists as three individual artifacts, but only one questionnaire was returned); (ii) the same editorial group (or
subsets of the same group)
is behind two or more ontologies; (iii) the ontology in question is being re-engineered and merged with one or more others; (iv) the ontology is no longer maintained (fortunately this is only true in one case). For these reasons the surveying process has also been quite challenging
, but this is a natural consequence of the dynamic nature of the ontology development process.



ADW is outdated,
questionnaire filled
for \'ETHAN’
, the new ontology

Drosophila development
, Drosophila gross anatomy, Fly taxonomy, FlyBase Controlled Vocabulary

.
Beside a textual evaluation we have visualized the main results in Table 2, showing responses to queries that were to be answered in a yes/no manner and Figure 1, illustrating qualitative survey responses to selected questions that were stated in an open manner.
Of the 42 survey respondents, 14 stated that they had developed their own naming conventions (question 2.1, Table 2). A closer look revealed that most of the documented conventions were both limited in coverage and embedded in papers or general style guides. 18 responders reported that in every case where naming conventions were documented, the naming of classes was tackled (question 2.2, Table 2). 10 responders stated they have conventions for naming relations, 9 had conventions on class IDs, 8 on namespaces, 7 on the name/version of the ontology itself and 8 responders had conventions for instance names. Of the respondents that did not document their naming conventions, 13 were using the ‘GO style guide’ for general guidance (see question 2.3, Figure 1). Both the Ontology for Biomedical Investigations (OBI) and Proteomics Standards Initiative (PSI) looked to the Metabolomics Standards Initiative (MSI) naming conventions document for guidance. Ontologies dealing with chemicals (e.g., ChEBI) usually draw on the International Union of Pure and Applied Chemistry (IUPAC) conventions for naming of small molecules. Of the species-specific ontologies, the Medaka Fish Ontology draws on the Zebrafish ontology; the mammalian phenotype group re-uses conventions of the Phenotype, Attribute and Trait Ontology (PATO). In the protein domain, InterPro and the Molecular Role Ontology (MRO) use UniProt names.

The survey further revealed that, many respondents captured more than one class name category, including synonymous alternative names, abbreviations and acronyms, brand names and plural forms (question 3.1, Figure 1). 31 respondents stated they capture a ‘user-preferred name’; 15 stated they capture a ‘formal name’ adhering to defined naming principles, 11 captured a short name to be displayed in large graphical representations; and 5 responders captured broader / narrower / related names. 5 respondents captured database cross-references (dbxref), and only 3 responders stated they capture foreign language translations in their artifacts.

The need of a number of ontology initiatives for higher-resolution representations of name categories, evidenced by the results described above is addressed explicitly in the survey (question 3.3, Table 2). It reveals that a third of respondents wished for a more detailed treatment of this issue, and for distinct representational units to be made available for the classification of these name categories in ontology languages themselves. Another third of respondents had no opinion on this issue, while the remainder - mostly OBO format users - felt that current naming practice is satisfactorily addressed in their current formalisms. It is notable that the problem of synonym types has been recognized by the Ontology Task Force of the W3C Semantic Web Health Care and Life Sciences Interest Group, which asserts that heterogeneous naming, and a lack of harmonization in the way names are represented in ontology languages, forces the construction of more complex queries by users and hinders the development of generic user interfaces.

The need for higher resolution in the representations of name categories is further illustrated by the development by many groups of their own metadata schemes to govern naming (e.g., OBI and MSI metadata annotation properties).