The range of inSubset should be an IRI, not a literal #1527
Comments
|
As of also
|
SELECT ?o (COUNT(?s) AS ?s_count)
(IF(isIRI(?o), "IRI", "Literal") AS ?type)
WHERE {
?s <http://www.geneontology.org/formats/oboInOwl#inSubset> ?o .
}
GROUP BY ?o |
PREFIX rdfs: <http://www.w3.org/2000/01/rdf-schema#>
SELECT ?s (GROUP_CONCAT(?c;
SEPARATOR=" | ") AS ?comments)
WHERE {
?s rdfs:subPropertyOf+ <http://www.geneontology.org/formats/oboInOwl#SubsetProperty> .
OPTIONAL {
?s rdfs:comment ?c
}
}
GROUP BY ?s
order by ?s |
|
|
I think I missed an important point when I re-opened this issue on December 17th I was complaining that the string-object in-subset statements were still present... but that's because I was looking at the most recent https://github.com/EnvironmentOntology/envo/blob/master/envo.owl, which doesn't contain the merged un unreleased contents of branch |
|
I do see this change from branch issue-1527 in envo-edit.owl in the main branch now: from to |
|
OK, this is actually in a very good state now except for these, as determined by src/envo/reports/envo-subsetTable.tsv
|
|
The build process is still able to create a fresh src/envo/subsets/envoPolar.tsv with ~ 670 rows. I still don't understand that process, since 'envoPolar' is mentioned by name in the definition of |
|
probably because I don't know anything about OWLTOOLS subsets/%.owl: subsets/envo-basic.obo
$(OWLTOOLS) $< --extract-ontology-subset --fill-gaps --subset $* -o $@.tmp && \
$(ROBOT) annotate -i $@.tmp --ontology-iri $(ONTBASE)/$@ --version-iri $(ONTBASE)/releases/$(TODAY)/$@ -a owl:versionInfo $(TODAY) \
--output $@ |
The range of inSubset should be an IRI, not a string literal. We have a mix at the moment
valid:
not valid:
These are the counts of invalids
environmental_hazards|90
envoAstro|72
envoAtmo|111
envoCesab|5
envoCloudAtlas|10
envoCmecs|36
envoCryo|40
envoEOVs|9
envoEmpo|37
envoMarine|62
envoMeo|27
envoNceas|22
envoOmics|35
envoPlastics|69
envoPolar|442
nlcd2011|22
subset_siren|37
wwfBiome|41
We already have IRIs defined for the subsets - nothing links to them so far
Note: this issue should not be used for discussing whether we use inSubset or a different AP. We have a different mega-ticket for this: #1202. Fixing the current ranges does not preclude using a different AP in the future, and would be a good incremental step towards this goal.
The text was updated successfully, but these errors were encountered: