Merging two classes

[cmungall]

(enhancement request)

OBO-Edit has a tool for merging two classes. This is described here:

http://oboedit.org/docs/html/Merging.htm

(this doesn't describe the effects of merging other axioms types, like equivalence axioms or disjoint axioms)

This is one of the features missed by OE users when I turn them to Protege; the steps must be performed manually.

A Protege merge tool would be more generic than the OE one - for example, Protege does not make any assumptions about specific annotation properties. I think OE users would be more than happy with a generic merge, i.e. essentially the same thing as using the EntityRenamer in the OWLAPI (with the source URI preserved somehow)

This functionality may belong in a plugin, e.g.
https://github.com/balhoff/obo-actions

(which performs the related 'obsoletion' operation)
cc: @dosumis

[matthewhorridge]

@cmungall has there been any progress on this elsewhere? Just wondering whether to assign this to the 5.6.0 milestone or not.

[cmungall]

Not really. We have a command in owltools: owlcollab/owltools@67c2aa1

Not worth reusing the code though

[matthewhorridge]

Ok, thanks, then I'll add it to the 5.6.0 milestone. (The plan is to get 5.0.0 asap, with just bug fixes for the 5.5.0 milestone).

[matthewhorridge]

p.s. assuming you still want this!

[cmungall]

Definitely! @ukemi do you have any particular preferences as to how this should operate?

[ukemi]

We should ping @pgaudet on this. She does the majority of merges in GO. I had envisioned a plug-in where the target term and the term-to-be-merged would be entered.

[matthewhorridge]

Just coming back to this @cmungall, @ukemi and @pgaudet

The link in the first comment on this issue is now broken, so I'm just wondering... If somebody was to merge term A into term B then this would mean that all occurrences of A are replaced with B right? What should happen to special annotation assertions such as rdfs:label and the like? Do these become synonyms?

[cmungall]

OboEdit would sometimes create a merged definition that you had to edit - but in other ontologies where people were less disciplined than GO, they would forget and you would see these kicking around...

in fact some still are:
https://www.ebi.ac.uk/ols/search?q=%22merged+definition%22&groupField=iri&start=0

I am OK with labels becoming exact syns in the OBO profile.

I think it's OK to temporarily have two text definitions and the curator to choose the desired one

We could also annotate any merged axioms with an annotation mergedFrom=<originalIRI>

[matthewhorridge]

Thanks @cmungall. Very helpful. Is there an IRI for "mergedFrom", or is this new?

[matthewhorridge]

Also, @cmungall, one thing that I'm wondering about, what's the main difference between a merge and a deprecation? If, say I merge X into Y, how is this different from deprecating X and replacing it with Y? I'm wondering this from both a technical point of view and from a process point of view.

[cmungall]

A merge is more extreme, and all that is left is a bare IRI (in fact in the obo, the stanza disappears altogether)

owlcs/owlapi#317

http://wiki.geneontology.org/index.php/Merging_Ontology_Terms

[pgaudet]

What is confusing perhaps is that the IRI of the merged term shows up (at least in protege) as if it was obsolete.

For example: GO_0004887 is merged (is a secondary ID for GO:0004879)
but is also has the 'owl deprecated = true' tag:

Pascale

[matthewhorridge]

Hi @pgaudet and @cmungall

I've implemented some merge functionality according to the discussion above and the GO editor's workflow docs. I've built a snapshot build of 5.6.0-beta-1, which can be obtained here:

macOS

Windows

Linux

Platform Independent

When you get a moment, please would you have a look at the merge functionality and let me know whether you're happy with it or whether it needs improvements (of any kind)?

You select a source entity and then choose "Edit -> Merge entity..." to select a target entity that the source entity will be merged into.

[pgaudet]

Hi @matthewhorridge This is great ! It seems to be working for me. I have not fully tested the process (ie merging into the real ontology), but the diff looks fine.

Thanks a lot !!!

Pascale

[matthewhorridge]

Hi @pgaudet Great! Thanks for testing if for me. I'll close this issue for now (as implemented), but please don't hesitate to let me know if this needs improvements.