Enriched mapping documentation #741
Conversation
cmungall
commented
Apr 25, 2024
•
cmungall
commented
- added a Commands/Mappings notebook
- extended the OAK mappings guide to talk about some difficulties around directionality
- clarified how weights work in lexmatch rules
- removed duplicate synonymizer section in rules data model (it now has its own data model)
Codecov ReportAttention: Patch coverage is
❗ Your organization needs to install the Codecov GitHub app to enable full functionality. Additional details and impacted files@@ Coverage Diff @@
## main #741 +/- ##
==========================================
- Coverage 75.54% 74.74% -0.81%
==========================================
Files 275 278 +3
Lines 32190 32758 +568
==========================================
+ Hits 24317 24484 +167
- Misses 7873 8274 +401 ☔ View full report in Codecov by Sentry. |
|
Is the Because when I try to use it directly on a OBO file, it gives a completely different output: $ runoak -i uberon.obo mappings UBERON:0000995
This ontology interface contains 267 prefixes:
AAO -> http://purl.obolibrary.org/obo/AAO_
ADO -> http://purl.obolibrary.org/obo/ADO_
ADW -> http://purl.obolibrary.org/obo/ADW_
AEO -> http://purl.obolibrary.org/obo/AEO_
[…]It goes on to list all the prefixes, but never actually lists any mapping. Whereas using it through the SQLite adapter gives the same results as in the doc: $ runoak -i sqlite:obo:uberon mappings UBERON:0000995
subject_id: UBERON:0000995
predicate_id: oio:hasDbXref
object_id: BTO:0001424
mapping_justification: semapv:UnspecifiedMatching
subject_label: uterus
subject_source: UBERON
object_source: BTO
[…]Is that a bug or the expected behaviour? If the latter, this warrants a mention in the doc. |
Co-authored-by: Damien Goutte-Gattat <[email protected]>
|
Hmm, on my phone now so can't check now but I believe that behavior is when pronto can't parse the obo |
| - Some mapping predicates such as ``skos:broadMatch`` and ``skos:narrowMatch`` are inherently directed, with the meaining inverted in the opposite direction | ||
| - Many mappings are only be asserted in one ontology, and may be "invisible" when queried from the mapped ontology | ||
|
|
||
| At the time of writing, most ontologies bundle mappings that are uncommitted `oio:hasDbXref`. |
There was a problem hiding this comment.
What does "bundle mappings" mean?
uncommitted?
Is the prefix oio mentioned earlier or 110% known by readers what this is?
|
|
||
| While mappings are often thought of as bidirectional, this is not quite true in practice: | ||
|
|
||
| - Some mapping predicates such as ``skos:broadMatch`` and ``skos:narrowMatch`` are inherently directed, with the meaining inverted in the opposite direction |
There was a problem hiding this comment.
directed --> directional?
this with the meaining inverted in the opposite direction may not be very clear and also just noticed typo in meaining
| While mappings are often thought of as bidirectional, this is not quite true in practice: | ||
|
|
||
| - Some mapping predicates such as ``skos:broadMatch`` and ``skos:narrowMatch`` are inherently directed, with the meaining inverted in the opposite direction | ||
| - Many mappings are only be asserted in one ontology, and may be "invisible" when queried from the mapped ontology |
There was a problem hiding this comment.
Might want to expand on why mappings may be "invisible" from an ontology, e.g. each ontology must agree to the mapping and include the mapping and actively assert the mapping to the term in the other ontology for it to be visible from either ontology.
| @@ -74,61 +80,43 @@ classes: | |||
| attributes: | |||
| subject_source_one_of: | |||
| multivalued: true | |||
| description: The source of the subject to be matched. | |||
| Multiple values can be provided, it must match at least one. | |||
There was a problem hiding this comment.
reading only from the gh diff, an example value looks like it would be useful to add
