Potentially dubious type 2 diseases

[cmungall]

For general background see #149

Briefly: At time t1, we have a disease D. At time t2, there is a purported discovery of a new type, which gets called D2. This effectively renders D a prototype.

The question then becomes how to interpret the semantics of D. The clearest thing to do is to make D generic and have an explicit subtype D1 (sib of D2). However, in many cases the purported D2 may turn out not be real, in which case we are left with a single-isa hierarchy with no difference between D1 and D. If D2 does turn out to be real it results in confusion over what D means, with some mentions in the literature presumably denoting D1. However, in these cases referring to D1 as D is not wrong, just lacks specificity.

This ticket is for collecting discussion on what type2s we believe to have insufficient evidence to warrant breaking down D. Check the box if we think insufficient evidence

• [x] Marfan type 2 #101
• [ ] Camurati-Engelmann type 2 #399
• [ ] Zimmermann-Laband Syndrome 2 https://omim.org/entry/616455 (Orphanet has no distinct entry)

[cmungall]

In #399 @pnrobinson suggests an n=1 tag. We could do this as a subset tag (inSubset in OWL).

If we do retain the n=1s, then I think the relationship to the generic form should perhaps be something other than SubClassOf, to avoid true path violations. analog_of, phenocopy_of, ...?

[nicolevasilevsky]

This issue was moved to monarch-initiative/mondo#162