Building a ISCN converter tool as a part of HGVS's suite of tools ?

[Laufer, Vincent]

Dear HGVS Nomenclature Team,

Thank you for the excellent work you do to empower so many clinical, research, and other enterprises that make use of molecular genetic information! I am writing to ask your advice on feedback on a planned tool.

Tool Being Considered: I have been considering building a tool to convert HGVS Nomenclature to ISCN, but would like to reach out to you before investing time in it.

Motivating Use-Case: The institution I work at (Michigan Medicine) is in the early stages of bringing up a WGS assay for children with a presumed genetic diagnosis. Because many pediatricians and other health professionals have long depending on cytogenetics and karyotyping in this context, I thought it might be judicious to include ISCN nomenclature in the final report, even though the core assay will be a next generation sequencing assay that will issue as output either a VCF or perhaps gVCF. 

As we plan to annotate any/all genetic variants using HGVS, I thought a conceptually clean approach might be to write a tool that converts HGVS nomenclature to ISCN nomenclature for a given (structural) variant. I looked at other available reports from academic medical centers who offer WGS as a service, and it appears that these institutions also do this. 

Does such a tool already exist? Before proceeding, I looked around as much as I could to see if such a tool already exists. In doing so, I found a few private companies that I think must have a tool like this - but I did not find an open-sourced tool.

I also checked HGVS's pages themselves, and found several relevant pages, e.g. 

      https://hgvs-nomenclature.org/stable/recommendations/DNA/complex/ and  

      https://hgvs-nomenclature.org/stable/software/ 

Here again some related tools exist, but I don't see a direct corollary ... So, I wanted to ask,

1. - Does this exist already somewhere I missed? 
   - Does HGVS have plans to tackle this, perhaps in collaboration with ISCN?
   - Do you have recommendations or advice I might learn from?
   - Is this an area of interest of yours?
   - Any other advice or notes on how to proceed, if you think the idea is meaningful?

Planned Approach: Unless the tool exists or the tool doesnt have merit, I was going to ask if this would be along the lines of a planned approach you might recommend or support: As per 2^nd link, 

1. Fork the hgvs-nomenclature repo
2. Create a new entry in the docs/software/ directory, using an existing entry as a template
3. Build the documentation locally (see README.md) to ensure that your software entry appears as intended
4. Submit the tool to a peer reviewed journal
5. Submit a pull request once done

Thanks very much for your advice and feedback,

Vincent Laufer, M.D., Ph.D. (he, him, his)

UM Experimental Pathology Physician-Scientist Training Track

Fellow in Molecular Genetics 2023-2024

Instructor in Clinical Informatics 2024-2025

vla...@med.umich.edu | Phone: (404) 594 9834  |  Pager: 24131

- I am a Compass Peer Supporter. Please feel free to reach out if you'd like someone to listen. 

- My working day may not be your working day. I respect your boundaries around personal time, well being, caretaking and rest. If you receive correspondence from me outside your working hours, please do not feel obliged to reply until you are at work.

**********************************************************
Electronic Mail is not secure, may not be read every day, and should not be used for urgent or sensitive issues

[Reece Hart]

Hi Vincent-

Converting HGVS to ISCN might be useful. I am unaware of any public tools that do this.

The HVNC maintains the specification, but does not maintain software. (Some of the members do this independently of their role on the committee.) 

Here are some recommendations:

• Be clear about the cases of HGVS expressions that you intend to be able to convert and those you don't. HGVS and ISCN have some overlap, but also have lots of unique capabilities. It will help you write tests and help users understand the scope for you to be clear about the conceptual overlap that you cover.
• Similarly, be clear about the clinically-relevent (to you or generally) ISCN expressions that aren't "reachable" (won't be generated) by your tool. For example, trisomy is not easily represented in HGVS. You will likely need to be able to report trisomies, but since you won't see them in VCF, you might need to figure out you'll generate them when appropriate. 
• Consider contributing to an existing tool rather than writing a new tool.

FWIW, I work for MyOme and we will need similar functionality shortly. We are likely to implement that in the Python hgvs library. If you are familiar with Python and interested in contributing to that code, please come see us at https://biocommons.org/. There are a few of us who can help you get started.

Regarding your "Planned Approach", you should write or contribute code and publish it before submitting a PR to the software list.

-Reece

[Laura Conlin]

Hi - I am on the ISCN committee, and we will have a lot of updates (a lot more examples, including trisomies) for structural variants in the next version due out in the fall, which I assume will also be associated with an HGVS website update as it was updated in collaboration with HGVS. For nomenclature for structural variants and copy number variants from sequencing technologies, the ISCN nomenclature is a combination of ISCN and HGVS, so I am not sure what you mean by converting between the two, unless you mean being able to write the ISCN portion of the combo nomenclature from just the HGVS part?  For instance, an unbalanced translocation using ISCN would be:

seq[GRCh38] der(9)t(9;10)(q34.3;q24.2)
NC_000009.12:g.137175878_qterdelins[NC_000010.11:g.99919751_qter]

With the first line stating the finding in ISCN karyotype-like nomenclature, and the bottom line with the HGVS. For something like this, there currently is no stand-alone ISCN for this type of finding, as the ISCN portion is more meant to supplement the HGVS portion (by sort of translating into a cytogenetic finding). Since this example doesn't have the sequence breakpoints in the ISCN-portion, it is not sufficient to describe the variant alone.

That being said, I don't know of any free tools that can do this (write the ISCN portion of a combined ISCN-HGVS nomenclature for structural variants). If you find one, I would be very interested to know about it!

Laura