monarch-initiative
MONDO:0011929 xref & def showing for all ORPHA terms in HPO browser
@matentzn helpfully brought me here! Sorry I'm not much help I originally thought this was a HPOA/HPO website issue.
Bug/Typo/Error description
Currently all ORPHA diseases display an incorrect MONDO xref to MONDO:0011929. I came across this browsing ORPHA term annotations on the HPO website.
The definition is the same (incorrect again from MONDO:0011929) for each ORPHA disease as well.
OMIM MONDO xrefs look good though! (At least browsing via HPO)
Originally posted:
TheJacksonLaboratory/hpo-web#289
Dear @peupeubangbang, Thank you for reaching out to the Mondo Disease Ontology, part of the Monarch Initiative!
We have received your ticket and will review it as soon as possible.
If you have any additional information or updates relevant to your inquiry, please add them in a comment on this ticket. Your input is valuable in helping us understand and resolve the issue effectively.
Thank you for your patience and for your contributions to Mondo!
Sincerely, The Mondo Team
The exact same xref was reported by another user on slack (@bryanlaraway), so it may be worth getting to the bottom of this.
@peupeubangbang can you add a link to the HPO browser for one of the HPO terms that displays this issue?
Here's the one I encountered first: https://hpo.jax.org/browse/disease/ORPHA:84 Specifically the issue is at the top of the disease view, the Mondo ID is indicated as MONDO:0011929 and the corresponding link is to the 'chromosome 1p36 deletion syndrome' page: https://monarchinitiative.org/MONDO:0011929
Thanks @bryanlaraway!
I don't see how this is an issue in Mondo. The ID for "Fanconi anemia" in Mondo is [MONDO:0019391](https://www.ebi.ac.uk/ols4/ontologies/mondo/classes/http%253A%252F%252Fpurl.obolibrary.org%252Fobo%252FMONDO_0019391) not MONDO:0011929 as shown on the HPO site and web service.
There are only 4 terms in Mondo with xrefs to Orphanet:84 (assuming that is what the HPO site is using this Orphanet ID) and these Mondo terms are all related to "Fanconi anemia".
@matentzn why do you think this is a problem with Mondo and not the HPO web site?
The HPO Annotations file does list Orpha:84 as being FA
ORPHA:84 Fanconi anemia HP:0001000 ORPHA:84 TAS HP:0040281 P ORPHA:orphadata[2024-04-24]
ORPHA:84 Fanconi anemia HP:0001053 ORPHA:84 TAS HP:0040281 P ORPHA:orphadata[2024-04-24]
ORPHA:84 Fanconi anemia HP:0001172 ORPHA:84 TAS HP:0040281 P ORPHA:orphadata[2024-04-24]
ORPHA:84 Fanconi anemia HP:0001871 ORPHA:84 TAS HP:0040281 P ORPHA:orphadata[2024-04-24]
ORPHA:84 Fanconi anemia HP:0001873 ORPHA:84 TAS HP:0040281 P ORPHA:orphadata[2024-04-24]
...
I asked Nico to review these findings and he said he contacted the HPO folks and it's because a wrong relationship on the HPO website (xref instead of skos:exactMatch) is being used. Closing this issue since it's not a Mondo issue.
Hi @peupeubangbang, Thanks for bringing this to our attention.
First and foremost, there was a nasty bug in the ontology annotation network (hpo's annotation api). That's been resolved and deployed.
Traditionally I'd use the monarch api to auto resolve the mondo identifier. But moving to OAN, I decided to do the mapping on load. I decided to use the base ontology xrefs knowing they would contain these relationships. However, it seems we are dropping "related" items in there as well. The temporary deployed fix does first a scanning for the xref match, which can return 0, 1, or many results. I then add a filter to see if the names start with eachother to narrow it down to one.
Moving forward a more permanent solution is to involve the sssom files into the data load for the mapping OR setup, configure and deploy the https://github.com/mapping-commons/sssom-api
https://hpo.jax.org/browse/disease/ORPHA:84
Along the way this helped identify other unrelated issues in phenol!
Thanks for the report! Closing since https://github.com/TheJacksonLaboratory/ontology-annotation-network/releases/tag/v1.0.7
