Species interaction type searches now use interaction vocabulary, do we need an actual ontology?

[globalbioticinteractions]

Species interaction type searches now use interaction vocabulary, do we need an actual ontology?

[reiz]

Being able to do short cypher queries is always worth a refactoring.

[cmungall]

In the OWL model (just checked in, eol-globi-ontology), the interaction class has different subclasses for each type of interaction. You wouldn't want to mirror this is java, but you might want to model the interaction hierarchy as nodes much like a species taxonomy. Thus you could do queries such as "find all symbiotic interactions" without explicitly listing all subtypes (mutualist, parasite), just as you can query for mammals without listing every mammal.

[jhpoelen]

@cmungall Thanks for modeling the interaction types as such. This addresses the original issue. I'll take a stab at using the ontology in an owl exporter as discussed. When would you like to share your results with Malcolm Storey? I am sure he'll have some valuable input. . .