Fix definition of 'sensory peripheral neuropathy' and 'motor peripheral neuropathy'

[allenbaron]

Is your request related to a specific disease? Please describe.
'sensory peripheral neuropathy' DOID:2491

Describe the proposed change(s)
It looks to me like the definition of 'sensory peripheral neuropathy' (DOID:2491) is insufficient:

A neuropathy that involves damage to nerves of the peripheral nervous system.

I propose updating the definition of 'sensory peripheral neuropathy' to change 'peripheral nervous system' to 'peripheral sensory nerves'?

---

For reference it's sibling 'motor peripheral neuropathy' (DOID:2477) has the definition:

A neuropathy that is characterized by inflammation or degeneration of the peripheral motor nerves.

[allenbaron]

Copied from Slack conversation with @lschriml & @sbello:

Lynn:

text definition looks OK to me
How would this change the axiom for uberon?

Allen:

There isn't any anatomy axiom on it. Uberon doesn't appear to represent the sensory peripheral nervous system at all.

Lynn:

you can use: peripheral nervous system, UBERON:0000010

Allen:

This is already an axiom on the grandparent of these terms 'peripheral nervous system disease' (DOID:574).

Sue:

Possibly you could use sensory nerve (UBERON:0001027) located in peripheral nervous system. The text definition could also be changed to '...sensory nerves of the peripheral nervous system'

Allen:

That's a good suggestion for wording.

Does the definition of UBERON's 'sensory nerve' cover all the nerves of the human peripheral sensory nervous system? Aren't internal sensory nerves included in the human system? UBERON:0001027 has the definition (emphasis added):

A nerve that transmits from sensory receptors on the surface of the body to the central nervous system.

Sue:

I think it should, if you look at the child terms it includes interoceptors that sense things like blood pressure and oxygen level. Possibly you could take surface to mean any surface, internal or external?

Allen:

Oh... I interpreted 'surface' to mean exposed surface (including gut, lung, etc) but I suppose this may be surface in a more abstract sense.

Sue:

Yeah, I had that same thought which was why I went to look at the child terms. I think it may just be an old and not well written definition.

Allen:

@lschriml, what do you think about adding 'disease has location' some 'sensory nerve'?

It would be nice to add an equivalent axiom to 'motor peripheral neuropathy' (DOID:2477) but it looks like UBERON doesn't have 'motor nerve'. It does have 'efferent nerve' (UBERON:0006798). Do you think that's close enough to be safe?

As an alternative we could use the better defined cell terms in CL: sensory neuron and motor neuron. (One note that this may not support all the desired reasoning since 'disease has location' still doesn't have a 'part of' property chain... so reasoners won't interpret these axioms to mean that these are also part of the 'peripheral nervous system').

That last comment will apply to any non-system term we use.

Lynn:

I wouldn’t use CL, how is efferent nerve defined?

Allen:

UBERON's efferent nerve is defined as you would expect for the term I think:

A nerve that transmits nerve impulses from the central nervous system to effectors such as muscles or glands.

It has motor nerve as a related synonym.

This is probably why neither of these diseases has a location axiom at the moment.

Allen:

We could make an axiom like:

'disease has location' some (
      'sensory neuron' AND
      'part of' some 'peripheral nervous system'
)

Kind of overkill though for the disease ontology to declare such an anatomy axiom though, if you ask me.

It would be nice to have an active (funded) human anatomy ontology.

Lynn:

Efferent does not sound like it would work for the peripheral system as the direction is toward the outside rather than from the edges (peripheral) back to the central system

We have not used 'part of' before to define location.  If the axiom does not use part of, I think it is ok

Allen:

The part of was mimicking CL's approach for some neuron locations. You said you want to avoid CL anyway (sorry, probably shouldn't have thrown that in there). Okay, here's what I'll do (let me know if you want a different approach).

1. I'm going to document this in an issue (so we can find this conversation highlighting the difficulty with location axioms... it was bigger than I expected)
2. I'm going to use sensory nerve (from UBERON) for sensory peripheral neuropathy and update the definition using Sue's suggestion)
3. I'll open an issue with UBERON and see if they'll add these two types of peripheral nervous systems or if they'll add motor nerve, since efferent nerve isn't really sufficient.

Lynn:

Great !! Good plan Allen !!

[allenbaron]

Opened a request at UBERON to add these terms: obophenotype/uberon#3653