tgbugs
commented
4 years ago The issue here is that we removed the classes that were the full intersection of all the phenotypes when we purged the intermediate classes (related: your question about why there are so many intermediate Markram classes). I can add these back if you need them.
I think the larger take home here is that we may need to clearly delimit the use cases that we are going to address in the paper, and more importantly, we should recognize that the surface level presentation of an ontology really does need to reflect a specific set of use cases. Nearly every issue that we are encountering now is the result of a decision we made earlier about how we wanted the ontology to appear. This suggests that we need a way to separate the data content of an ontology from its presentation -- but not rendering as a web page, rendering as another ontology. This is precisely what the spec for the InterLex export functionality can enable once it is implemented, and the troubles we are encountering now will provide concrete use cases that can be used to guide that work.
You aren't including the morphological phenotypes in the name as you are for other classifications. I need this for some of the figures. It used to be there.