NCI-Thesaurus / thesaurus-obo-edition

OBO Library edition of NCIt
22 stars 8 forks source link

Peripheral neuropathy vs Neuropathy #20

Open cmungall opened 6 years ago

cmungall commented 6 years ago

see the two *** classes:

 / NCIT:C7057 ! Disease, Disorder or Finding
  is_a NCIT:C2991 ! Disease or Disorder
   is_a NCIT:C27551 ! Disorder by Site
    is_a NCIT:C26835 ! Nervous System Disorder
     is_a NCIT:C27580 ! Peripheral Nervous System Disorder
      is_a NCIT:C4731 ! Neuropathy *** 
       is_a NCIT:C116381 ! Neuritis
       is_a NCIT:C119734 ! Peripheral Neuropathy ***
       is_a NCIT:C125389 ! Small Fiber Neuropathy
       is_a NCIT:C26748 ! Diabetic Neuropathy
       is_a NCIT:C26951 ! Polyneuropathy
       is_a NCIT:C27025 ! Ischemic Neuropathy
       is_a NCIT:C27033 ! Autonomic Neuropathy
       is_a NCIT:C27055 ! Uremic Neuropathy
       is_a NCIT:C27187 ! Inherited Neuropathy
       is_a NCIT:C27221 ! Entrapment Neuropathy
       is_a NCIT:C27301 ! Axonal Neuropathy
       is_a NCIT:C27338 ! Congenital Neuropathy
       is_a NCIT:C27595 ! Femoral Neuropathy
       is_a NCIT:C27596 ! Peroneal Neuropathy
       is_a NCIT:C28155 ! Horner Syndrome
       is_a NCIT:C35098 ! Idiopathic Neuropathy
       is_a NCIT:C70938 ! Mononeuritis Multiplex
       is_a NCIT:C75467 ! Charcot-Marie-Tooth Disease
       is_a NCIT:C84553 ! Amyloid Neuropathy
       is_a NCIT:C84728 ! Giant Axonal Neuropathy
       is_a NCIT:C84950 ! Optic Neuritis

examining the defs:

the distinction can be justified as the parent is more inclusive in including the cranial "nerves" which include for example cranial nerve II which is not a true nerve. However, the parent "Neuropathy" is already includes under PNS disorder, which with a naive axiomatization will force Neuropathy and Peripheral Neuropathy to be equivalent.

I also note that some of the direct children of Neuropathy such as CMT should really be classified as peripheral neuropathies? An exception would be optic neuritis which affects CNII, which is not a true PNS nerve.

I'm not sure what the best solution is here. I see the optic "nerve" exceptions causing these kinds of fudges in different ontologies and I'm unsure of the best solution that satisfies both logic and normal terminological use, see for example https://github.com/obophenotype/uberon/issues/298

NCIEVS commented 6 years ago

From the editing group:

"We moved neuropathy directly under nervous system disorder. We did not put the term peripheral neuropathy under non-neoplastic peripheral nervous system disorder because cancer can definitely cause peripheral neuropathy. We double-treed peripheral neuropathy under neuropathy and peripheral nervous system disorder as well. I think that this resolves the issue."

cmungall commented 6 years ago

Thanks! I think your reasoning makes sense here, I agree about cancer potentially causing peripheral neuropathy.

I wasn't very clear with my somewhat wandering question however. To be more direct: what would happen if we were to merge Neuropathy + Peripheral Neuropathy? Based on their definitions (affects PNS/CNs vs affects PNS), the main distinction is that the former includes the cranial nerves. But the cranial nerves are all PNS!

The problem here is the awkward CN II which is not really a true PNS nerve. I was wondering if CN II is the sole reason for the distinction or if there was something I was missing.

But this isn't really very high priority, it's more of a discussion point, I will close if there is no action here.