In-frame insertion including stop codon not following updated HGVS

[gaberudy]

The following variant: NM_001123385.1:c.2574_2575insGTTCCTTAGGAG introduces a stop codon and according to the updated HGVS spec I believe should be represented as p.E858_L859insValProTer)

Currently: https://mutalyzer.nl/name-checker?description=NM_001123385.1%3Ac.2574_2575insGTTCCTTAGGAG

It produces NM_001123385.1(BCOR_i001):p.(Leu859_Trp1755delinsValPro)

The updated HGVS page describes this case with an example: https://varnomen.hgvs.org/recommendations/protein/variant/insertion/

insertions containing a translation stop codon in the inserted sequence are described as an insertion, not as a deletion-insertion removing the entire C-terminal amino acid sequence

p.(Met3_His4insGlyTer) the predicted consequence at the protein level of an insertion at the DNA level (c.9_10insGGGTAG) is the insertion of GlyTer (alternatively Gly*) NOTE: this is not described as p.(Met3_Ile3418delinsGly), a deletion-insertion replacing the entire C-terminal protein coding sequence downstream of Met3 with a Gly)

This is definitely an edge case, but I do think the new HGVS recommendation

[jfjlaros]

Mutalyzer makes its protein description by comparing the reference protein sequence to the observed sequence. This is the only way to make two variants that result in the same protein have the same description.

We realise that this is not completely in line with the current HGVS recommendations, but pending a fixed set of rules to resolve this issue, we have no way of dealing with this in a structured way.