After refactoring the subclass of sync, a lot of changes

[matentzn]

Build PR for #8431

I eyeballed some of the, and they seem correct to me.. To review this:

1. Check out this file: https://raw.githubusercontent.com/monarch-initiative/mondo-ingest/main/src/ontology/reports/sync-subClassOf.confirmed.tsv (this is what is merged into mondo-edit)
2. Confirm that the changes reflect the changes you see to the edit file
3. Confirm that the changes to the sync-subClassOf.confirmed.tsv are intended

The most important is (3), as I already did 1 and a few spotchecks re 2.

I suggest @joeflack4 you take a look here is well.

[twhetzel]

I'm not sure I am looking at this in the "right" direction. For example, for MONDO:0000022 it has these changes:

- is_a: MONDO:0024290 {source="NCIT:C118172", source="icd11.foundation:1048673005"} ! enuresis
+ is_a: MONDO:0024290 {source="icd11.foundation:1048673005"} ! enuresis

So in Mondo MONDO:0000022 'nocturnal enuresis' is a subclassof MONDO:0024290 'enuresis'. The Mondo term has xref: icd11.foundation:1048673005 {source="MONDO:equivalentTo"} and icd11.foundation has this same subclassof relation between 'nocturnal enuresis' and 'enuresis' and is a source for subclassof relationship in Mondo. But the Mondo term also has xref: NCIT:C118172 {source="MONDO:equivalentTo"} and the subclassof relationship between between 'nocturnal enuresis' and 'enuresis' exists in NCIT, but NCIT:C118172 was removed as a source for the subclassof relationship in Mondo.

Also, many of the subclassof sources that were removed are from DOID and NCIT. When I looked in the mondo-ingest repo at tmp/component-download-ncit.owl.owl I could find 'nocturnal enuresis' and 'enuresis', but when I looked at components/ncit.owl I could not find either class 'nocturnal enuresis' and 'enuresis'.

[matentzn]

Also, many of the subclassof sources that were removed are from DOID and NCIT. When I looked in the mondo-ingest repo at tmp/component-download-ncit.owl.owl I could find 'nocturnal enuresis' and 'enuresis', but when I looked at components/ncit.owl I could not find either class 'nocturnal enuresis' and 'enuresis'

At first I found this concerning, but then I checked our specs and they clearly state that we only ever sync the neoplasm branch with NCIT. enuresis is in the psychiatric disorder branch..

The DO is more interesting/concerning:

[Term]
id: MONDO:0000248
name: dengue shock syndrome
xref: DOID:0050125 {source="MONDO:equivalentTo"}
is_a: MONDO:0005502 {source="DOID:0050125", source="MONDO:Redundant"} ! dengue disease

[Term]
id: MONDO:0005502
name: dengue disease
xref: DOID:12205 {source="MONDO:equivalentTo", source="EFO:0005547"}

And in DO:

<owl:Class rdf:about="http://purl.obolibrary.org/obo/DOID_0050125">
    <rdfs:subClassOf rdf:resource="http://purl.obolibrary.org/obo/DOID_12205"/>
    <rdfs:label xml:lang="en">dengue shock syndrome</rdfs:label>
</owl:Class>

The confirmed subclass table is clearly missing this, only containing these two entries related to dengue:

MONDO:0005358	Dengue hemorrhagic fever	MONDO:0005502	DOID:12206	DOID:12205	dengue disease
MONDO:0000259	asymptomatic dengue	MONDO:0005502	DOID:0050143	DOID:12205	dengue disease

In my opinion its a bug in the subclass sync, one that we didn't notice because my previous subclass pipeline didn't delete all the evidence..

So, I think we need to:

• 1. File a bug report (if you agree it is one) (@twhetzel)
• 2. Fix that bug (@joeflack4)
• 3. Merge asap into main (maybe we should considering ditching the whole idea of a develop branch, its slowing us down a lot) (@twhetzel)
• 4. Make a new data release (@twhetzel)
• 5. Run the subclass pipeline Fix subclass of sync #8431 again (@matentzn)
• 6. Update this PR again (@matentzn)
• 7. Review again (@twhetzel)

let me know what you think

[twhetzel]

I added the ticket monarch-initiative/mondo-ingest#708, unclear what availability @joeflack4 has to fix this as an urgent priority. No, we're generally not ditching the develop branch. BUT in this specific case for the December release may need to make some careful adjustments to get these bug fixes addressed.

[twhetzel]

Converting to Draft until this is fixed monarch-initiative/mondo-ingest#708

[joeflack4]

General review

I spot checked some things. Looks like I'd expect; the deletions and lack thereof.

[joeflack4]

Double checking the is_a source

Just using the example case from monarch-initiative/mondo-ingest#708, even though that is an direct-source-indirect-mondo and not part of this PR.

subject_mondo_id	subject_mondo_label	object_mondo_id	subject_source_id	object_source_id	object_mondo_label
ID		SC %	>A oboInOwl:source
MONDO:0000248	dengue shock syndrome	MONDO:0005502	DOID:0050125	DOID:12205	dengue disease

id: MONDO:0000248
name: dengue shock syndrome
xref: DOID:0050125 {source="MONDO:equivalentTo"}
is_a: MONDO:0005502 {source="DOID:0050125", source="MONDO:Redundant"} ! dengue disease

I'm pretty sure source="DOID:0050125" is correct but I just wanted to double check that this shouldn't be `source="DOID:12205", because in a way I feel like it makes sense to express the evidence that way as well.

[joeflack4]

-added?

The PR title just says "subclass sync", but your comments are only about -confirmed, so that's what I checked. Have we not implemented -added for the subclass sync, or has it just not been run for this PR?

[joeflack4]

Confirm that the changes to the sync-subClassOf.confirmed.tsv are intended

@matentzn What are you asking precisely? I don't think you mean to the TSV; Like I don't think you're asking me to go to mondo-ingest and look at the diff of that file over time.

It sounds like what you're asking is that the changes by the TSV are intended.

By intended, it sounds like you want to take this time to double check the mondo-ingest subclass sync pipeline itself; to spot check a few cases where we can match up what we'd expect to see appear in the -confirmed TSV, and then check to see if those entries do indeed appear in the TSV. Is that what you are asking?

I don't really have any doubts about this, but I did do a couple checks:

• Checked a random SCR that wasn't removed by this PR, verified the matches in mondo.sssom.tsv exist, and checked that the component .owl does corroborate the SCR.
• Checked a couple cases like this: A deletion, (MONDO:0017626 is_a: MONDO:0018100 {source="Orphanet:306522"} ! familial primary hypomagnesemia. Checked and saw that that MONDO:0018100 has no Orphanet mapping in mondo.sssom.tsv, so makes sense that this one was removed.
• I wanted to check a case like: Mondo is_a evidence for a source was removed, and exist in the mondo.sssom.tsv, but it was only removed because the SCR didn't exist in the source, but it's best if I don't spend time digging around for such a case unless requested.