HTAN ID Minting Exploration and Scoping

[aclayton555]

24-11/12 Sprint Kick-Off Discussion:

HTAN Identifiers - do these really need to be human readable

• NO! Provenance can be established by another identifier (e.g. UUID)
• Do we make a case here to make changes? This work has been supported out of DFCI, so this group would need to take this on
• We could have centers onboard to these as-is, then just stop using these later? Could get tricky with publications though
• But the current identifiers and implementing these is burdensome on us and centers
• Make no changes for now, but perhaps consider ID minting in Y8. Is this something ISB could support? Was previously in research proposal

TL:DR - Dar'ya will investigate potential options for mint IDs and if ISB can support this.

[aclayton555]

24-11/12 Close-Out:

Dar’ya will need to dig into this a bit more and understand if/how this can be done, how it can be resourced, etc.
Some time sensitivity here as IDs will be subject of upcoming info sessions. Need to think about messaging here and what an MVP could look like...what are the priority immediate steps rather than a full solution

Dar’ya to create a one-pager to more tightly scope this, considering:

• What do we (HTAN DCC) need?
• What do submitters need?
• What do re-users need?

Expect this to continue into next sprint.

[PozhidayevaDarya]

Linking draft: https://docs.google.com/document/d/1yJ10hroGty8pOcuIQVJNjXlyHK3y-u3Gv_DcPiicsx8/edit?tab=t.0

[aclayton555]

Discussed on 2025.01.06 HTAN DCC Coordination call:

• In general, preference for conservative approach to any changes to identifiers, however, recognized that there are opportunities to improve on this from HTAN1. If we are going to make any changes though, need to decide on these ASAP.
• Consider whether we actually need to make any changes to the identifiers themselves OR is it preferred to provide improved guidance to centers on how to implement the existing approach?

25-1 Kick-off:

• Distinguish between biospecimen (e.g. start with a 0 in the last integer block) and data file ID (e.g. start with a 1 in last integer block)
• Need to review latest biospecimen RFC and understand levels of samples are are needed to be captured here (sample, sub-sample, assayed sample) and what we are enforcing here.
• Specific guidelines on pooled and TMA assays. This is particularly confusing where we have one data file that represents data from multiple participants. Why have Participant IDs in data file metadata - this can be inferred from the provenance
• Identifiers for control and blank samples (this was not straightforward in HTAN1)

Dar'ya will put together a presentation with proposed approaches for discussion on upcoming HTAN DCC call.

[aclayton555]

2025.01

• Dar'ya put together proposal (https://docs.google.com/presentation/d/11K0iC8zUrOemOTKIdsDWQgWN1hsmmM9KrK3MIIF8z30/edit?usp=sharing), which has been discussed at the HTAN DCC level, and includes HTAN contributor feedback from OHSU and WUSTL.
• Ethan's guidance (see Slack): we don't need a full-blown RFC for the ID changes. But, Alex and I agreed we need a DCC-specific RFC for ID changes. This way, all DCC reps can vote on that too, but since the changes are fairly constrained, we don't need a full community RFC.

@PozhidayevaDarya I think we are ready to close out this particular ticket and create a separate ticket for the RFC itself?

[PozhidayevaDarya]

@aclayton555 Yep I agree and that sounds good to me!