add negative hpo phenotypes

[nlwashington]

we currently only import the "abnormal" phenotypes, but we also need to bring in the "normal" ones.

we also need to set up a general model for these kinds of statements.

@cmungall please comment on the proper association structure.

[mellybelly]

There are also the opposite phenotypes curated by @micheldumontier, would be great to get those in

[drseb]

There are also the opposite phenotypes curated by @micheldumontier, would be great to get those in

Is there more information on this available somehere?

— Reply to this email directly or view it on GitHub.

[micheldumontier]

Yes, using a set of paired antonym modifiers,

'hyper' =>'hypo', 'increased' =>'decreased', 'elevated' =>'depressed', 'enlarged' =>'reduced', 'elevation' =>'depression', 'prolonged' =>'shortened', 'high' =>'low', 'positive' =>'negative', 'stenosis' =>'dilation', 'thick' =>'thin', 'presence' =>'absence', 'large' =>'small', 'prominent' =>'receded', 'long' =>'short'

I found that 2308 terms contain one of these modifiers. I was able to match 302 pairs with existing HP terms. So, there may be the possibility to add more antonym terms to HP.

[micheldumontier]

https://docs.google.com/spreadsheets/d/1VTh1wzujK0q-Kejn1671KjX3IYiKKlfeiJYuaMfVSic/edit?usp=sharing

[cmungall]

Awesome! I will check these against the ones we get from the DL axioms in PATO. (inc/dec in magnitude axioms)

[cmungall]

Instead of the class C, use the expression

[owl:complement_of C]