Expand the SFWO with the Bibliographic Ontology

[Archilegt]

We will use the Bibliographic Ontology in the EUdaphobase Taxonomy Ontology, the Chilopoda Anatomy Ontology, and the Diplopoda Anatomy Ontology. I recommend the same for the SFWO.
Additional resources:
Bibliographic Ontology Specification
Bibliographic-Ontology-BIBO on GitHub

Part of what I am doing for the SFWO is to expand existing terms with additional definitions and sources. That is why it is important to formalize the structure of the sources.

BIBO example for a standard journal article:
The article is directly linked to its journal. The ordered author list is represented using an rdf:List.

<info:doi/10.1134/S0003683806040089> a bibo:Article ;
    dc:title "Effect of argillaceous minerals on the growth of phosphate-mobilizing bacteria Bacillus subtilis"@en ;
    dc:date "2006-01-01" ;
    dc:isPartOf <urn:issn:23346587> ;
    bibo:volume "42" ;
    bibo:issue "4" ;
    bibo:pageStart "388" ;
    bibo:pageEnd "391" ;
    dc:creator <http://examples.net/contributors/2> ;
    dc:creator <http://examples.net/contributors/1> ;
    bibo:authorList ( <http://examples.net/contributors/2> <http://examples.net/contributors/1>) .

<urn:issn:23346587> a bibo:Journal ;
    dc:title "Applied Biochemistry and Microbiology"@en ;
    bibo:shortTitle "App Biochem and Biol"@en .

Note that if the BIBO is added to the SFWO, then the SFWO instructions manual, part "Ontology design" should be updated to reflect the change.

[nleguillarme]

So it seems it's not possible to use the BIBO ontology to annotate OWL classes.

In the Gene Ontology, the source for the definition is embedded in the definition itself, for instance with a link towards a web page (http://amigo.geneontology.org/amigo/term/GO:0033058)

I also asked OMO people if there exists some constructs to specify alternative definitions for a given term (see information-artifact-ontology/ontology-metadata#113)

[Archilegt]

The notation [GOC:mtg_MIT_16mar07] is quite cryptic and doesn't point to anything else that can be recognized as a proper source. In this case the source seems not to be a publication but the Gene Ontology Consortium (GOC) itself. The rest of the string, apart from the date, may refer to mitochondrial genome (mtg_MIT). We need another example. But how incredible that such an important ontology has bad quality "source" strings.

[Archilegt]

I recommend exploring the Hymenoptera Anatomy Ontology (HAO). Here is the term transepisternal line, which contains elaborated references in the field "Label usage (sensu)" and in two "alternative definition". However, I could not see those sources while exploring the term in the BioPortal. Maybe I am missing something, or maybe the citation problem has never been solved.

References:
Yoder MJ, Mikó I, Seltmann KC, Bertone MA, Deans AR (2010) A Gross Anatomy Ontology for Hymenoptera. PLOS ONE 5(12): e15991. https://doi.org/10.1371/journal.pone.0015991
Seltmann et al. (2012) A hymenopterists’ guide to the Hymenoptera Anatomy Ontology: utility, clarification, and future directions. Journal of Hymenoptera Research 27: 67-88. https://doi.org/10.3897/jhr.27.2961

Meanwhile, in the Spider Anatomy Ontology they also seem to include references within the concept: "Definition/Concept: For example, “most posterior appendage of prosoma,” followed by a bibliographic reference; sometimes the name of a curator replaces the reference."

Reference:
Ramírez, Martín J., and Peter Michalik. 2019. "The Spider Anatomy Ontology (SPD)—A Versatile Tool to Link Anatomy with Cross-Disciplinary Data" Diversity 11, no. 10: 202. https://doi.org/10.3390/d11100202

Overall, this does not look good. Definitions are definitions, without a "tail" composed by a reference, and structured and resolvable references are important. We should try finding a solution for this.

Back to BIBO, if it is not possible to use it in its RDF form, couldn't we just transform it to another format? E.g., from RDF to OWL? Or couldn't we use any other "bib" ontology from the BioPortal? E.g.: BIBFRAME has many classes that we could reuse but I don't know if it has all, and it doesn't seem to have the structured format from BIBO that is so useful for articles.

The other option is to build a small "RIS Format Ontology" using this wiki as a source for the terms and definitions. That would have the advantage that we could model it as needed, plus I understand the format because I frequently use it. Would you like to try that first, @nleguillarme?

[nleguillarme]

See information-artifact-ontology/ontology-metadata#113