Stream: genomics
Topic: Update on TBD Code Systems for FHIR
Rachel Kutner (Apr 27 2020 at 17:55):
This page has a list of TBD Code Systems. (http://build.fhir.org/ig/HL7/genomics-reporting/tbd-codes.html#tbd-codes-functional-annotation) Epic is looking into representing some of these concepts in our data structure, and we'd like to use a version of the HL7v2 spec to send and receive them. Is there an update on when these concepts will be submitted and reviewed for LOINC approval? We'd like to be able to reference the LOINC codes if they're available. We're particularly interested in the Functional Annotation and Mode of Inheritance concepts.
Kevin Power (Apr 27 2020 at 19:30):
I would suggest for those you are interested in right now, can you log a tracker for our workgroup to submit a request to LOINC? @Liz Amos has offered to help get our requests vetted and routed to LOINC.
Bret H (Apr 28 2020 at 15:40):
@Rachel Kutner do you happen to have a prioritized list? Last count, I think there's 5 codes outstanding. Also, I think the delay is in part due to needing clear useful definitions ratified by CG WG vote. Rachel, do you happen to have ideas on clarifying the definitions? It would be a great help. Many thanks!
Rachel Kutner (Apr 28 2020 at 15:52):
@Kevin Power I'm unfamiliar with what a tracker is or how to log it, but if I understand what to do I certainly can help do that.
@Bret H Thanks - I'm meeting with my coworker this afternoon to discuss this more and will get a list of higher priority terms from the link and ask if we have any ideas for clarifying the definitions.
I work on the Genomics Team at Epic and we have relationships with several clinical genomics groups around the country who are invested in improving interoperability, resulting, representation, etc. of genomic results in the clinical space. If there's ever need for clarification on terms or clinical input, feel free to pull me in and I can work to involve our community members who are excited about providing feedback and input.
Kevin Power (Apr 28 2020 at 16:03):
@Rachel Kutner - A 'tracker' is a change request. HL7 uses JIRA now to track those change requests, so you can log a change request here: https://jira.hl7.org/ (you will need to request an account). If you have questions as you log it, let me know.
Rachel Kutner (Apr 28 2020 at 17:58):
@Bret H We're mainly interested in Mode of Inheritance and Functional Annotation.
For Mode of Inheritance, there's a LOINC code suggested in the link I originally included, but the LOINC's answer list isn't fully inclusive of the terms we're wanting to create. Would the best approach be to submit a request to update the LOINC code to include the full list of terms we're interested in? The list is here: https://ftp.ncbi.nlm.nih.gov/pub/GTR/standard_terms/Mode_of_inheritance.txt
Or is there some reason a new LOINC code might be warranted if we want to expand the scope?
For Functional Annotation, we have a few questions. Based on the description of the term in the link, it points to a SO code which indicates Structural_Variant, which seems incorrect. http://sequenceontology.org/browser/current_svn/term/SO:0001537
It possibly intended to reference the Functional Effect list here, which corresponds to our concept of "Functional Effect": http://sequenceontology.org/browser/current_svn/term/SO:0001536
If it did intend to reference the Structural_Variant SO, that is what we're considering "Variant Consequence" in our own data structure: https://m.ensembl.org/info/genome/variation/prediction/predicted_data.html
Which term is intended to be referenced by "Functional Annotation"?
Is the WG considering including one or both of these concepts (our 'Variant Consequence' and 'Functional Effect') in future Variant Specs, either for FHIR or HL7v2?
Kevin Power (Apr 28 2020 at 18:58):
Rachel Kutner said:
For Mode of Inheritance, there's a LOINC code suggested in the link I originally included, but the LOINC's answer list isn't fully inclusive of the terms we're wanting to create. Would the best approach be to submit a request to update the LOINC code to include the full list of terms we're interested in? The list is here: https://ftp.ncbi.nlm.nih.gov/pub/GTR/standard_terms/Mode_of_inheritance.txt
Or is there some reason a new LOINC code might be warranted if we want to expand the scope?
If the LOINC code is otherwise OK (name, description, etc.. are all OK) but you just want additional items in the answer, that is probably a question for LOINC. You are welcome to have this group review the additions.
Kevin Power (Apr 28 2020 at 19:05):
Rachel Kutner said:
For Functional Annotation, we have a few questions. Based on the description of the term in the link, it points to a SO code which indicates Structural_Variant, which seems incorrect. http://sequenceontology.org/browser/current_svn/term/SO:0001537
It possibly intended to reference the Functional Effect list here, which corresponds to our concept of "Functional Effect": http://sequenceontology.org/browser/current_svn/term/SO:0001536
If it did intend to reference the Structural_Variant SO, that is what we're considering "Variant Consequence" in our own data structure: https://m.ensembl.org/info/genome/variation/prediction/predicted_data.htmlWhich term is intended to be referenced by "Functional Annotation"?
I don't have time now to dig back into the full history of this, but here is the tracker/JIRA that made the request: J#22815 -- Perhaps @Patrick Werner or @Jamie Jones can recall more details?
Kevin Power (Apr 28 2020 at 19:07):
Rachel Kutner said:
Is the WG considering including one or both of these concepts (our 'Variant Consequence' and 'Functional Effect') in future Variant Specs, either for FHIR or HL7v2?
Most all of what we discuss here is for the FHIR IG. However, we can always consider updates to V2 as well if/when needed. If we come to the conclusion that a V2 update is needed, I would have to find out how log requests for V2 LRI changes - I have never done that before.
Liz Amos (Apr 28 2020 at 20:30):
@Rachel Kutner I just want to clarify that 79742-3 Inheritance pattern based on family history is what you had in mind for "mode of inheritance"? That specific code is not currently part of the v2.5.1 spec but it is in the CG FHIR profile. If so, I will summarize your proposals/questions to Clem and he can help figure out the best approach to help facilitate the LOINC side of things.
Kevin Power (Apr 29 2020 at 20:19):
Kevin Power said:
Rachel Kutner said:
Is the WG considering including one or both of these concepts (our 'Variant Consequence' and 'Functional Effect') in future Variant Specs, either for FHIR or HL7v2?
Most all of what we discuss here is for the FHIR IG. However, we can always consider updates to V2 as well if/when needed. If we come to the conclusion that a V2 update is needed, I would have to find out how log requests for V2 LRI changes - I have never done that before.
@Rachel Kutner - I think you can log requests for the V2 LRI here: http://www.hl7.org/dstucomments/showdetail.cfm?dstuid=226
Arthur Hermann (May 04 2020 at 16:00):
@Kevin Power - One of the issues that I believe may be a problem for users of EPIC (~60% or more of health plans) is that they built their genomics module on an earlier version of the STU - and they are working in HL7 v2. Since a very large number of health care systems will eventually be impacted by this, I believe we should discuss this during one of our meetings. @Rachel Kutner - please correct me if I am wrong. This becomes a significant issue if these organizations plan on having labs send results directly into Epic.
Jamie Jones (May 04 2020 at 16:08):
Our IG does have some v2 mapping functionality, though it's obviously not a 1-1 situation. We should update the mapping for next release and could open a tracker to increase guidance there, if nothing else. Long-term, we need to sync with the 2-to-FHIR project.
Kevin Power (May 04 2020 at 17:57):
@Arthur Hermann -- The link I provided to @Rachel Kutner is where requests for the V2 LRI can be logged. If similar changes need to be made to the FHIR IG, that is requested and tracked via JIRA which you can request here: https://jira.hl7.org/ (making sure to select "Specification" as Genomics Reporting (FHIR)
Arthur Hermann (May 04 2020 at 18:00):
@Kevin Power @Rachel Kutner - this may be a bigger issue than just logging a few items... Kaiser is meeting with Epic in 2 minutes to discuss, their genomics module. I will see what I can find out. We also have 1 to 2 Epic employees consistently attend our WG. I believe they should be bringing up any concerns as Rachel has... this would make more sense to me (if our Epic friends agree).
Kevin Power (May 04 2020 at 18:12):
@Arthur Hermann - Are you referring to a mapping from the V2 LRI to our FHIR IG? As Jamie said, we have some basic things, but could have more and we would welcome someone providing that sort of input for STU2. As Jamie also mentioned, there is a separate project going that is contemplating a very broad v2-to-FHIR mapping, and given we have capacity, we will need to reconcile that work as well.
To be clear, my earlier comments were more procedural. Rachel has asked some good questions, and we as a work group will need to decide how to appropriately answer those questions. When the answer is "we need to update our specs," then we need a request logged that our group can formally consider and vote on.
Rachel Kutner (May 04 2020 at 18:18):
@Arthur Hermann I'm not sure if it's clear - I'm one of the Epic employees that consistently attends the WG. I'm working to put together a call to get more information and coordinate with you on our end. We'll be reaching out soon to coordinate!
@Kevin Power That's great to hear - are there resources you can point me to wrt what mapping has already been accomplished? I'd also appreciate being involved in any future conversations regarding mapping, as this is something I've thought about a fair amount.
Kevin Power (May 04 2020 at 18:29):
@Rachel Kutner - Not much I'm afraid. I was thinking we did more with this page, but we haven't (yet): http://build.fhir.org/ig/HL7/genomics-reporting/v2.html -- it basically has nothing in it.
Another place to look is here: http://build.fhir.org/ig/HL7/genomics-reporting/variant-mappings.html ... it has some mappings built, but I will admit I haven't vetted it. Our other profiles have a mapping tab as well, but the variant one is probably the best place to start.
Arthur Hermann (May 04 2020 at 18:48):
@Kevin Power thank you so much for the clarification - my point is that we need to discuss exactly the issues you state above to determine the best path forward
Arthur Hermann (May 04 2020 at 19:10):
@Rachel Kutner @Kevin Power - i would be very happy to speak to you Rachel as you stated(and I did know you were one of the Epic people who attends the WG :) . The only issue is that speaking to Epic is very political at KP - so I want to make sure we clear this with the appropriate people. Kevin - more than happy to have you on board as well!
Rachel Kutner (May 12 2020 at 16:03):
Hi all, thanks for the discussion today during the call. What did we want to do as a follow up on this?
Patrick Werner (May 12 2020 at 16:07):
Hi @Rachel Kutner thanks for brining this topic up. I created a new dedicated topic for this Discussion:
https://chat.fhir.org/#narrow/stream/179197-genomics/topic/Functional.20Annotation.20.2F.20Variant.20Consequence.20.2F.20Functional.20Eff
Jamie Jones (May 12 2020 at 16:08):
4 concepts on there are taken care of and can be removed: mode-of-inheritance and the 3 pgx effects (other than high-risk allele, which needs thought)
Jamie Jones (May 12 2020 at 16:10):
region-coverage, variant-inheritance, diagnostic-implication, and therapeutic-implication need a caretaker to curate a writeup and submit to LOINC
Jamie Jones (May 12 2020 at 16:11):
exact-start-end, inner-start-end, outer-start-end need updating on the original loincs if possible.
Jamie Jones (May 12 2020 at 16:11):
that leaves functional-annotation and associated-cancer, which are undergoing active discussion on calls
Bret H (May 14 2020 at 18:11):
Jamie Jones said:
region-coverage, variant-inheritance, diagnostic-implication, and therapeutic-implication need a caretaker to curate a writeup and submit to LOINC
@Jamie Jones I can take it. but need to be certain no one else is. I do not want to submit a duplicate LOINC request
Kevin Power (May 14 2020 at 18:12):
Only other person who might have started something would be @Liz Amos
Bret H (May 14 2020 at 18:37):
I sent the LOINC code (94186-4) to the group for : variant-inheritance Variant inheritance A quality inhering in a variant by virtue of its origin. The terms are in the sequence ontology under SO:0001762. LOINC responded a while back. CompletedTermReport_Submission-5169_2.xlsx
Bret H (May 14 2020 at 18:39):
image.png 94186-4 Origin of germline genetic variant
Bret H (May 14 2020 at 18:40):
it is not appearing in the build. Can it be added now? It was created by LOINC specifically for the component.
Liz Amos (May 14 2020 at 20:54):
@Jamie Jones I think these will need requests (with definitions!). I checked at the data type is range for those codes, so it might not make sense to have exact.
Liz Amos (May 14 2020 at 20:55):
Alas, I have not put anything together. I keep getting on other projects. I highly recommend coming up with a clear definition (ideally with a source of the definition) of what the concept means and then use the template provided by LOINC to share with the group.
Last updated: Apr 12 2022 at 19:14 UTC
