Why are there near-identical disease terms in the Platform (ADHD), and how to collate the evidence?

[Anita]

There are two similar terms related to “ADHD” in the Platform, one hyphenated, one not:

However, the search results such as the number of targets and profiles of them are different. For example, for term “attention deficit hyperactivity disorder”: 1617 targets without any animal models recording; but for term “attention deficit-hyperactivity disorder”: 2427 targets and most of them have animal models recordings. For the same gene (such as DRD2), evidences of the same disorder are different. This seems strange.

So, my questions are:

1. I’m confused about the reasons of the differences.

2. What’s more, what should I do to find all evidences of ADHD in your platform?

3. Maybe I can download two tables.tsv of these terms then combine them, but how to deal with the differences of “overall association score” for the same gene?

These information provided by Open Target Platform are important for my research projects. Thanks for providing the useful tool of Open Targets Platform. It’s convenient for users to search all evidences of interested genes and diseases.

[hcornu]

Hello!

Thank you for your question; I’m glad to hear that the Platform is useful for your research.

I have opened a ticket so that our data team can investigate why we have two near-identical terms for ADHD in the Platform:

Near identical terms for ADHD in the Platform

In the Platform, attention deficit hyperactivity disorder (EFO_0003888) is listed as a child term of attention deficity-hyperactivity disorder (MONDO_0007743):

This means that evidence for EFO_0003888 is considered indirect evidence for MONDO_0007743. When searching for targets associated with MONDO_0007743 in the Platform, or displaying all evidence for the association of a target with MONDO_0007743, this will also display evidence from the child term, EFO_0003888.

In other words, you should find all evidence for ADHD by searching for attention deficity-hyperactivity disorder. You can query this in the GraphQL API, or use our data downloads direct and indirect evidence files.

You can find more information about this in our documentation: Target - disease associations - Open Targets Platform Documentation

I hope this helps!

Best wishes,

Helena

[Anita]

Hi, Helena,

Thank you so much for posting and dealing with this. I’ve seen all the informaiton you mentioned above, that’s clear and helpful.

Only one thing I still want to make sure. As you said “displaying all evidence for the association of a target with MONDO_0007743, this will also display evidence from the child term, EFO_0003888.”, does it mean the evidence of the child term will also display in the MONDO_0007743? I think it should be yes, but the Bibliography of MONDO_0007743 only contains 12 publications.

Could you pls explain that ? Thanks a lot.

Best,
Anita

[hcornu]

Welcome to the Open Targets Community, @Anita

Indirect evidence will only be displayed in associations pages and associations tables where the disease is the fixed entity.

So for example, if I look at the text mining evidence from EuropePMC for the association between DRD2 and attention deficit-hyperactivity disorder (MONDO_0007743), I will also see evidence for the association of DRD2 and attention deficit hyperactivity disorder (EFO_0003888), such as the paper “Attention-deficit-hyperactivity disorder and reward deficiency syndrome.” (Blum et al. 2008).

Disease profile pages will only show the data we have for that specific disease.

Does that answer your question?

[Anita]

Thank you Helena . I’m still confused, sorry to re-ask again.

My question is: why the number of the Bibliography for the two terms are so different? I think the Bibliography for attention deficit hyperactivity disorder should include in the Bibliography for attention deficit-hyperactivity disorder. Maybe I’m wrong. Pls check me

[hcornu]

Hi Anita!

EFO_0003888 and MONDO_0007743 are the same disease, but due to a bug, they are considered different diseases in the Platform.

Disease profile pages only display the evidence which we have linked to that specific disease. They do not display evidence from child terms.

Therefore, the bibliography for each term is separate, as they are considered separate diseases.

[Anita]

Finally, I understood Thank you for your kind reply, Helena. Looking forward to the update version of it.

[ochoa]

Related to this previous topic:

Duplicated diseases in 22.06

Due to a problem in the ontology, a number of rare diseases are duplicated in the 22.06. EFO is currently undergoing a cleanup of the Orphanet terms to assign them to their corresponding MONDO id when appropriate. As part of this process, a large number of terms have been duplicated in the ontology. [Screenshot 2022-07-12 at 10.48.41] We are working with EFO to resolve this issue ASAP:

[hcornu]

Hello!
This issue has been resolved in our latest update — 22.09