[Story] Explore the integration of EDAM ontology to OC

[tschaffter]

What product(s) is this story for?

OpenChallenges

As a user, I want

As a Challenge Organizer, I want my challenges to be annotated with existing ontologies so that the annotations are useful to a wider group of users.

As an OC Developer, I want to leverage existing ontologies as much as possible so that we don't reinvent the wheel.

Description

Explore how the EDAM ontology could be used to annotate challenge resources.

Acceptance criteria

No response

Tasks

No response

Anything else?

I think that the first time the EDAM ontology was suggested in the context of OC was for assigning values about the input data type.

Have you linked this story to a GitHub Project?

• [X] I have linked this story to a GitHub Project and set its metadata.

[gaiaandreoletti]

• [X] EDAM ontology added to OC excel file

[tschaffter]

EDAM ontology added to OC excel file

We should probably discuss what the integration would have looked like before updating the "production" spreadsheet. :slightly_smiling_face:

@gaiaandreoletti Can you please describe in this ticket what the solution you propose is?

[gaiaandreoletti]

To make it easier for me to search for the appropriate ontology based on the challenge data, I added the EDAM ontologies as a new column. However, we can decide to delete this column later on. Currently, the tab called 'input_data_types' contains the ontologies that correspond to the type of data found across all challenges listed in the excel file. Previously, this tab was empty. A drop down menu listing all ontologies was made created to facilitate process. iF the EDAM onthology for the type of data used in the challange was not available , I used either the topic or the operation reflecting the expetiment used. e.g.

Preotein function prediction operation_1777 Preoteomic experiment topic_3520 Pretein interaction data data_0906 Protein structure data_1460 Protein variants topic_3120 RNA sequencing topic_3170

[tschaffter]

I have never used EDAM personally but here are some questions:

• From the list you shared and from EDAM README page, I see that there are different concepts like operation, format, data. Are they all suitable to describe challenge input data type?
• Should the challenge input data type property take a single value or an array of values?
• Given the public information about a challenge, how would the protocol applied to identify the value(s) for the challenge input data type?

[tschaffter]

I think we have managed to narrow down how to annotate challenges with EDAM values!

Are we settling on what follows?

• Set input data type to an array of EDAM values (data, format and possibly topic ids)
  • Challenge may include multiple data type like, e.g. mammogram images and demographic information
• Add the property challenge.operation as a string that takes an EDAM operation id

@gaiaandreoletti @vpchung

[vpchung]

Set input data type to an array of EDAM values (data, format and possibly topic ids)

By array, did you mean a comma-separated list...? If so, I think it will be easier to follow the same table format as contribution_role where each row is a challenge_id and a edam_id (or edam_name). For example:

id	challenge_id	edam_name
1	1	Protein variants
2	2	Systems biology
3	2	Electronic health record

[vpchung]

Add the property challenge.operation as a string that takes an EDAM operation id

There are only 3 challenges so far with an operation annotation, so this is done! ✅

[tschaffter]

Update

@gaiaandreoletti presented the EDAM ontology to the team and received positive feedback. The implementation is currently underway.