MONDO, HPO and Orphanet

Data Access
1

Dear OT team,

I am wondering are we able to download all MONDO, HPO and Orphanet ID and clinical term from opentarget database？ or we need to come to each database to identify ID and term table?

···
In order to maximise the list of available disease - phenotype links, the Platform ingests data from the Monarch Merged Disease Ontology (MONDO) and the Human Phenotype Ontology (HPO), the latter in a joint effort with Orphanet.
···

Thanks,

Shicheng

2

Hi Shicheng,

Can I check — are you looking to download a list of all disease/phenotype terms in the Open Targets Platform with their corresponding IDs?

Or are you looking to download a list of phenotype IDs with their corresponding MONDO, HPO, Orphanet IDs? e.g. for lung disease, EFO_0003818, MeSH D008171, NCIt C3198, MedDRA 10051054, MONDO: 0005275

Best wishes,

Helena

1 Like
3

Dear Helena,

I hope this message finds you well. I am trying to understand whether all phenotypes from the HPO, MONDO, and Orphanet ID databases are included in the Open Targets Platform phenotype database. It is crucial for my current workflow that no phenotypes from HPO, MONDO, or Orphanet ID are missing.

Can I fully rely on the Open Targets Platform to include all these terms?

Thank you,

Shicheng

4

Hi Shicheng,

Not all phenotypes from HPO, MONDO, and Orphanet are included in the Open Targets Platform.

Our disease index is pulled from EFO, which has terms imported from other ontologies, but we do not pick up the entirety of HPO or MONDO.

5

Dear Helena,

I hope this email finds you well.

I’ve been exploring whether all phenotypes from the HPO and MONDO are included in the Open Targets Platform phenotype database.

So I readed the Platform infrastructure section of the OpenTargets documentation, specifically the code related to data preparation for the Extract, Transform, Load (ETL) pipelines ( platform-input-support ). In this code, I noticed that OpenTargets starts by downloading links from every database. For the disease/phenotype portion, the jq command is used to extract relevant information from OWL files, followed by conversion to EFO IDs. However, I couldn’t find any filters applied to Disease/phenotype data beyond null checks.

I would like to ask if OpenTargets applies any filters to phenotypes from HPO, MONDO, and Orphanet. If so, at what stage in the process does this filtering occur?

Best regards,
Hongyue

6

Hi @HongyueLi, and welcome to the Open Targets Community! :tada:

You can take a look at the the script that takes all those inputs to build the disease index in our ETL code: platform-etl-backend/src/main/scala/io/opentargets/etl/backend/Disease.scala at master · opentargets/platform-etl-backend · GitHub

1 Like
7

Dear Helena,

Thank you for the warm welcome and for pointing me towards the relevant script. :smiling_face:

Best,
Hongyue