logical def for dysmotility terms

[nicolevasilevsky]

@pnrobinson on the phenotype editors call today, we discussed the logical defs for these terms, which currently follow the pattern below:
HP_0002579 Gastrointestinal dysmotility
HP_0012850 'Small intestinal dysmotility'

equivalentTo:
text: "'has_part' some ('contractility' and ('inheres_in' some %s) and ('has_modifier' some 'abnormal'))"
vars:
-anatomical_entity

Instead of using the PATO quality 'contractility' in the logical def, do you think we could use GO terms:

HP_0002579 Gastrointestinal dysmotility

'has part' some
('decreased rate'
and ('inheres in' some 'gastro-intestinal system smooth muscle contraction')
and ('has modifier' some abnormal))

HP_0012850 'Small intestinal dysmotility'

'has part' some
('decreased rate'
and ('inheres in' some 'small intestine smooth muscle contraction')
and ('has modifier' some abnormal))

[nicolevasilevsky]

cc @matentzn

[pnrobinson]

The GO-based definitions seem reasonable. I guess the question is whether they will work in terms of subsumption inference for Q/C of HP?

[matentzn]

Yes, thanks @pnrobinson - We are keeping in an eye on these things. For now, the important bit is to determine common patterns for all cases and apply them consistently - contractility is such a specialised concept that I do not expect any changes in subsumption (maybe @nicolevasilevsky when you make the change, run the reasoner before and after and see what the reasoner says - this may be good practice whenever you change a pattern. What to look out for: new inferred super classes; new inferred subclasses).