Simple in frame deletions involving the stop codon

[Matt Ducar]

Can you help clarify which of the recommendations apply to these kinds of variants:

The genomic event: NC_000007.13:g.140434397_140434402del deletes the last coding codon and the stop codon of the transcript:  NM_004333.6:c.2296_2301del

It isn't clear to me what rules should apply here at the protein level.  NP_004324.2:p.H766_*767del doesn't seem correct as we are lengthening the amino acid sequence.

We have:  https://hgvs-nomenclature.org/stable/recommendations/protein/deletion/
variants extending the amino acid sequence at the C-terminal end with one or more amino acids, are described as an Extension.

However the syntax of an extension doesn't work as it needs to look similar p.*110Gext*17  where the  description must start with p.* and can not have an amino acid range.  

The closest rule seems to be under: https://hgvs-nomenclature.org/stable/recommendations/protein/delins/
Even though this isn't a deletion-insertion, it seems to be in the spirit of:
deletion-insertion variants on the DNA or RNA level that start N-terminal (5') of and including the translation termination (stop) codon, are described as frameshift.

So I'd lean towards a description similar to: NP_004324.2:p.H766fsN*?   where we might be able to calculate the value for ? but I'm not for this toy example.

But when I enter this variant into online resources the results don't make sense:
Variant Validator and Mutalyzer both describe it as a delins at the protein level
NC_000007.13(NP_004324.2):p.(His766delinsAsnLys)

Where I see what they are doing as continuing to translate into the UTR results in a Asn-Lys-Ter amino acid sequence.  But this doesn't seem correct as what would we do if there weren't a stop codon in the UTR?

Ensembl's Variant Effect Predictor (for GRCh37) doesn't seem to conform with any recommendations and reports:
NP_004324.2:p.His766_Ter767delextTer1

Any guidance is greatly appreciated.

[Johan den Dunnen]

Dear Matt,

the variant described, NC_000007.13:g.140434397_140434402del (NM_004333.4:c.2296_2301del), can not be described as an "extension variant" since the last amino acid of the predicted variant protein is changed. Following HGVS nomenclature, see https://hgvs-nomenclature.org/stable/recommendations/protein/deletion/, where it states "deletions on the DNA or RNA level that encode an open reading frame which after the deleted sequence shifts to another reading frame, are described as a frameshift" the variant should be described as a frame shift like p.(His766AsnfsTer3).

I agree the predicted consequences given by tools like VariantValidator and Mutalyzer, i.e. p.(His766delinsAsnLys), seem correct as well, but I see no basis for these based on the current HGVS nomenclature.

Best regards,

Johan den Dunnen
GVsharedLOVD-team

Op donderdag 3 oktober 2024 om 16:58:41 UTC+2 schreef matt....@gmail.com:

[Matt Ducar]

Dear Johan,

Thank you for the detailed answer.  Would a 3bp deletion of just the stop codon get described as an "extension" or a "frameshift"?  My intuition is "extension" since the entirety of the original amino acid sequence is maintained.  

Based on your answer in the last e-mail & my question above, may I suggest a clarification to your "notes" section on the "Protein -> Deletion" page:

It wasn't immediately obvious to a naive reader that a deletion that included the stop codon was considered a shift to a different reading frame.  With your explanation above, it now makes sense to me.   Having language like you have on the Deletion-Insertion page could clarify for future readers:  "Deletion events that start N-terminal (5') of and including the translation termination (stop) codon, are described as frameshift."  

 

It might also be helpful to clarify the classification of a deletion of the stop codon as others could have the same question.

Thanks again for all of your help.

[Johan den Dunnen]

Dear Matt,

indeed, following HGVS noemnclature, a 3bp deletion of just the stop codon should be described as an "extension.

Regarding frame shifts, maybe except for an extension, any shift to a reading frame that is not used in the reference sequence is considered as a frame shift.

Best regards,

Johan den Dunnen

HUGO HGVS Variant Nomenclature Committee (HVNC)

Op maandag 28 oktober 2024 om 20:18:09 UTC+1 schreef matt....@gmail.com: