Relation between "clavicle bone", "shoulder bone" and "shoulder joint" #1227

Closed
fbastian opened this Issue Jun 15, 2016 · 2 comments

Comments

Projects
None yet
2 participants
@fbastian

fbastian commented Jun 15, 2016

There are relations that lead to infer that "clavicle bone" exists only in tetrapoda, see: BgeeDB/anatomical-similarity-annotations#12 (comment)

Is this relation correct?
UBERON_0003461 "shoulder bone" has_part part_of UBERON_0016884 "shoulder joint"
And what is this relation "has_part part_of"?

OWL file:

<owl:Class rdf:about="http://purl.obolibrary.org/obo/UBERON_0003461">
    <rdfs:subClassOf>
        <owl:Restriction>
            <owl:onProperty rdf:resource="http://purl.obolibrary.org/obo/BFO_0000051"/>
            <owl:someValuesFrom>
                <owl:Restriction>
                    <owl:onProperty rdf:resource="http://purl.obolibrary.org/obo/BFO_0000050"/>
                    <owl:someValuesFrom rdf:resource="http://purl.obolibrary.org/obo/UBERON_0016884"/>
                </owl:Restriction>
            </owl:someValuesFrom>
        </owl:Restriction>
    </rdfs:subClassOf>
...
</owl:Class>

@fbastian fbastian referenced this issue in BgeeDB/anatomical-similarity-annotations Jun 15, 2016

Open

Release 2015-09-22 #12

@fbastian fbastian changed the title from Relation between "shoulder bone" and "shoulder joint" to Relation between "clavicle bone", "shoulder bone" and "shoulder joint" Jun 15, 2016

@cmungall

This comment has been minimized.

Show comment
Hide comment
@cmungall

cmungall Jun 15, 2016

Member

OK, a few things going on here

The 'source' axiom is:

'shoulder bone' EquivalentTo 'bone element' and overlaps some 'shoulder
joint'

(let's leave aside the state-specific bone and state-generic 'bone
element' for the moment)

Formally

overlaps = has_part o part_of

(ie if two things have a part in common, they overlap_

but we can't state this as an OWL axiom, so we expand as a
meta-reasoning step

Giving us:

'shoulder bone' EquivalentTo 'bone element' and has-part some (part-of
some 'shoulder joint')

Again ignore the bone/element mismatch, I think this is justifiable,
albeit circular (the def of shoulder joint references shoulder!)

There are some interesting discussions here about logic and how to
approach generic-sounding names like 'shoulder' that have a somewhat
fixed meaning...

but all that aside, the real issue here is:

'clavicle bone' SubClassOf 'shoulder bone'

This should clearly by a taxon GCI, for the reasons you mention.

And on top of that: why didn't the extensive taxon constraints catch
this?

Member

cmungall commented Jun 15, 2016

OK, a few things going on here

The 'source' axiom is:

'shoulder bone' EquivalentTo 'bone element' and overlaps some 'shoulder
joint'

(let's leave aside the state-specific bone and state-generic 'bone
element' for the moment)

Formally

overlaps = has_part o part_of

(ie if two things have a part in common, they overlap_

but we can't state this as an OWL axiom, so we expand as a
meta-reasoning step

Giving us:

'shoulder bone' EquivalentTo 'bone element' and has-part some (part-of
some 'shoulder joint')

Again ignore the bone/element mismatch, I think this is justifiable,
albeit circular (the def of shoulder joint references shoulder!)

There are some interesting discussions here about logic and how to
approach generic-sounding names like 'shoulder' that have a somewhat
fixed meaning...

but all that aside, the real issue here is:

'clavicle bone' SubClassOf 'shoulder bone'

This should clearly by a taxon GCI, for the reasons you mention.

And on top of that: why didn't the extensive taxon constraints catch
this?

@cmungall cmungall closed this in eec9bbc Jun 15, 2016

@fbastian

This comment has been minimized.

Show comment
Hide comment
@fbastian

fbastian Jun 16, 2016

Thanks for the explanations, and for the fix.

Thanks for the explanations, and for the fix.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment