Closed IanDMedeiros closed 2 months ago
Ahh, I see. Ecoregions are intended to have place-names (unlike most EnvO classes)
They just don't show up very nicely in OLS because they are named individuals, not subclasses.
Here's a pivot table of the current ecoregion properties. That gives some sense of other properties that could be used for these.
@IanDMedeiros can you share your ORCID, so that you can get credit as the creator? I'm not sure yet whether we should use the dc:creator
or oboInOwl:created_by
predicate.
You can also chose form a couple of different synonym styles if you want, and you could assert that any one of these ecpregions are part of (obo:BFO_0000050
) another.
These would be the first ecoregions in over 5 years, and ecoregion
has this comment:
ENVO's sister project, GAZ, contains terms for instances of ecoregions (e.g. Beringia lowland tundra). Requests for new terms should be directed to GAZ. ENVO will only contain this top-level class.
@pbuttigieg , @cmungall what guidance do you ahve?
Hi @IanDMedeiros @turbomam
OntoBee and OLS don't show individuals / instances well if at all.
@turbomam - that guidance on using GAZ is obsolete, as GAZ is not maintained. ENVO won't include hundreds of thousands of placenames, but it does make sense for it to include individuals for things like ecoregions, ecotones, biomes etc. Could you update that comment to:
Specific ecoregions are geographically defined and instantiated entities (e.g. Beringia lowland tundra), and should be represented as instances / named individuals of ENVO:ecoregion or an appropriate subclass.
@turbomam - thank you for adding these. Here's an example of how we represent a WWF ecoregion. They can indeed have asserted parthood, but be a little cautious on the domain/range constraints of relations, as some are not recomemnded for instances.
<!-- http://purl.obolibrary.org/obo/ENVO_01001569 -->
<owl:NamedIndividual rdf:about="http://purl.obolibrary.org/obo/ENVO_01001569">
<rdf:type rdf:resource="http://purl.obolibrary.org/obo/ENVO_01001570"/>
<obo:BFO_0000050 rdf:resource="http://purl.obolibrary.org/obo/ENVO_01001571"/>
<oboInOwl:created_by rdf:resource="https://orcid.org/0000-0002-4366-3088"/>
<oboInOwl:creation_date rdf:datatype="http://www.w3.org/2001/XMLSchema#dateTime">2019-03-05T17:25:21Z</oboInOwl:creation_date>
<oboInOwl:hasBroadSynonym>Western Australia Ecoregion</oboInOwl:hasBroadSynonym>
<oboInOwl:hasDbXref>WWF:AA1310</oboInOwl:hasDbXref>
<oboInOwl:hasDbXref>https://www.worldwildlife.org/ecoregions/aa1310</oboInOwl:hasDbXref>
<rdfs:label xml:lang="en">Western Australian Mulga Shrublands Ecoregion</rdfs:label>
</owl:NamedIndividual>
PS: we can also be more relaxed with the definitions of instances - they don't have to read as formally as class definitions. In fact it's also okay to omit them at the instance level. We can and should link out (as shown above) to the WWF, IUCN, or other source page for further description of the instance.
PPS:
I'm not sure yet whether we should use the dc:creator or oboInOwl:created_by predicate.
we have a mix currently, which we should normalise. I'm for the dc:creator as this is more broadly used.
Great feedback, @pbuttigieg. You can see the new issues you suggested here mentioned in this thread now.
@turbomam Here is the information you requested: https://orcid.org/0000-0003-2179-0745
Thank you!
example ecoregion instance:
Indo-Malay Ecoregion, ENVO:01001626
# Individual: <http://purl.obolibrary.org/obo/ENVO_01001626> (Indo-Malay Ecoregion)
AnnotationAssertion(<http://purl.org/dc/elements/1.1/creator> <http://purl.obolibrary.org/obo/ENVO_01001626> <https://orcid.org/0000-0002-4366-3088>)
AnnotationAssertion(<http://www.geneontology.org/formats/oboInOwl#creation_date> <http://purl.obolibrary.org/obo/ENVO_01001626> "2019-04-26T23:38:50Z"^^xsd:dateTime)
AnnotationAssertion(<http://www.geneontology.org/formats/oboInOwl#hasDbXref> <http://purl.obolibrary.org/obo/ENVO_01001626> <https://www.worldwildlife.org/biomes/deserts-and-xeric-shrublands>)
AnnotationAssertion(rdfs:label <http://purl.obolibrary.org/obo/ENVO_01001626> "Indo-Malay Ecoregion"@en)
ClassAssertion(<http://purl.obolibrary.org/obo/ENVO_01001570> <http://purl.obolibrary.org/obo/ENVO_01001626>)
Where does that fit into src/envo/envo-idranges.owl
?
It looks like it came from
Datatype: idrange:2
Annotations:
allocatedto: "Pier Luigi Buttigieg"
EquivalentTo:
xsd:integer[>= 1000000 , <= 1999999]
I will follow @pbuttigieg's example and use identifiers form my range for this. I'm happy to change that if necessary.
Also, I will be adding these directly into src/envo/envo-edit.owl
using these predicates suggested by @matentzn
and these
rdfs:label@en
rdfs:subClassOf
(from an RDF perspective)Sounds good @turbomam - stay close to the form of existing ecoregion instances if there's no compelling reason to vary
From https://www.worldwildlife.org/ecoregions/at1009 in red
Please note: These biome and ecoregion pages (and associated data) are no longer being updated and may now be out of date. These pages and data exist for historical reference only. For updated bioregion data, please visit One Earth.
@IanDMedeiros please see https://github.com/EnvironmentOntology/envo/pull/1526/files (or open the envo-edit.owl from this branch in Protege)
I am requesting new terms for several southern African ecoregions. These would be useful for annotating DNA sequencing data submitted to NCBI. EDITED: Added several additional term requests.
Proposed term: Lowland Fynbos and Renosterveld ecoregion Position in hierarchy: child of terrestrial ecoregion [ENVO:01001570] Definition: Area of the Cape Floristic Region of South Africa containing vegetation of the mediterranean shrubland biome [ENVO:01000217] occurring below 300 m. Database link: https://www.worldwildlife.org/ecoregions/at1202
Proposed term: Montane Fynbos and Renosterveld ecoregion Position in hierarchy: child of terrestrial ecoregion [ENVO:01001570] Definition: Area of the Cape Floristic Region of South Africa containing vegetation of the mediterranean shrubland biome [ENVO:01000217] occurring above 300 m. Database link: https://www.worldwildlife.org/ecoregions/at1203
Proposed term: Albany Thicket ecoregion Position in hierarchy: child of terrestrial ecoregion [ENVO:01001570] Definition: Area in the Eastern Cape of South Africa with vegetation characterized by sclerophyllous and succulent shrubs Database link: https://www.worldwildlife.org/ecoregions/at1201
Proposed term: Drakensberg Montane Grassland, Woodland, and Forest ecoregion Position in hierarchy: child of terrestrial ecoregion [ENVO:01001570] Definition: Area of the montane grassland biome [ENVO:01001808] in the Drakensberg mountains of southern Africa between 1800 and 2500 m. Database link: https://www.worldwildlife.org/ecoregions/at1004
Proposed term: Drakensberg Alti-Montane Grassland and Woodland ecoregion Position in hierarchy: child of terrestrial ecoregion [ENVO:01001570] Definition: Area of the montane grassland biome [ENVO:01001808] in the Drakensberg mountains of southern Africa above 2500 m. Database link: https://www.worldwildlife.org/ecoregions/at1003
Proposed term: Highveld Grassland ecoregion Position in hierarchy: child of terrestrial ecoregion [ENVO:01001570] Definition: Area of the montane grassland biome [ENVO:01001808] in the interior plateau of southern Africa between 1400 and 1800 m. Database link: https://www.worldwildlife.org/ecoregions/at1009