Interesting failure when creating a protein description

[ifokkema]

Describe the bug
I ran into an interesting issue while handling a variant, which seems to be linked to the predicted protein change.

When handling NC_000022.10:g.19929250_19929251insCCCCGCC, first leaving checkonly to the tx value, not requesting a protein prediction (link), the output is as expected. However, when setting checkonly to False (link), the output becomes:

{
  "error": "NP_006431.2:p.(Val26_Lys158delinsGlyArgGlyAlaGlyArgGlyAlaGlyArgSerSerArgSerAlaGlyLeuTerSerProGlyGlyArgArgGlyIleTrpTrpProGlyLeuCysGlnGlyGlyArgProAlaGlyLysGluGlyGlyArgGlyGlyLeuArgGlyThrPheSerProArgHisProValGlyProArgArgHisLeuArgGlnArgGlyLeuHisProGlnGluAlaAspAlaProGlyGlyThrAlaGlyArgProAspProArgCysProGlnLeuTrpLeuGlyGlyGlyProAlaArgAlaAlaTerLeuGluGluAspGlyArgSerCysSerLysSerArgGluIleLeuGluLeuGlyProProCysProAlaSerGlyGlnLysSerGlnValLeuTer): char 87: expected the character ')'"
}

The error suggests a problem with the generated protein description. In fact, there is, since this should be regarded as a frameshift variant and not a deletion-insertion. Besides, the inserted sequence contains several stop codons, which may be causing the error. I'm curious if this is perhaps related to the recent changes regarding the selenoproteins?

To Reproduce
See links above.

Expected behavior
Besides that the variant should be a frameshift on protein level, I expect a full API output instead of this error.

Thanks!

[Peter-J-Freeman]

This relates to

#815

and will be corrected once we merge and upload to the system (note it is not Selenon, but the fixes have resolved the p. predictions)

        "hgvs_predicted_protein_consequence": {
            "lrg_slr": "LRG_417p1:p.V26Gfs*132",
            "lrg_tlr": "LRG_417p1:p.Val26GlyfsTer132",
            "slr": "NP_006431.2:p.V26Gfs*132",
            "tlr": "NP_006431.2:p.Val26GlyfsTer132"
        },

[Peter-J-Freeman]

@ifokkema , can you confirm these outputs are correct though. Once this is approved, I can rebuild locally and confirm the response the API would give

[ifokkema]

Hmm... I would need to check the DNA to be sure, but if the protein description in the error in my report provides a correct amino acid insertion, then that fsTer132 would not be correct. I would also simply expect a Ter much earlier after a frameshift.

[Peter-J-Freeman]

for reference, the translations I have are

RefSeq: MAAMAVALRGLGGRFRWRTQAVAGGVRGAARGAAAGQRDYDLLVVGGGSGGLACAKEAAQLGRKVAVVDYVEPSPQGTRWGLGGTCVNVGCIPKKLMHQAALLGGLIQDAPNYGWEVAQPVPHDWRKMAEAVQNHVKSLNWGHRVQLQDRKVKYFNIKASFVDEHTVCGVAKGGKEILLSADHIIIATGGRPRYPTHIEGALEYGITSDDIFWLKESPGKTLVVGASYVALECAGFLTGIGLDTTIMMRSIPLRGFDQQMSSMVIEHMASHGTRFLRGCAPSRVRRLPDGQLQVTWEDSTTGKEDTGTFDTVLWAIGRVPDTRSLNLEKAGVDTSPDTQKILVDSREATSVPHIYAIGDVVEGRPELTPIAIMAGRLLVQRLFGGSSDLMDYDNVPTTVFTPLEYGCVGLSEEEAVARHGQEHVEVYHAHYKPLEFTVAGRDASQCYVKMVCLREPPQLVLGLHFLGPNAGEVTQGFALGIKCGASYAQVMRTVGIHPTCSEEVVKLRISKRSGLDPTVTGCUG*

VarSeq: MAAMAVALRGLGGRFRWRTQAVAGGGRGAGRGA[GRSSRSAGLUSPGGRRGIWWPGLCQGGRPAGKEGGRGGLRGTFSPRHPVGPRRHLRQRGLHPQEADAPGGTAGRPDPRCPQLWLGGGPARAAULEEDGRSCSKSREILELGPPCPASGQKSQVL*]

I put the square brackets in to show the changed sequence

[Peter-J-Freeman]

but if the protein description in the error in my report provides a correct amino acid insertion

It may not. I have now modified the code because of potential errors

[Peter-J-Freeman]

For reference. The HGVS Python library was previously used for insertion translations. It keeps messing up, so I have now passed it into our own translating code. This is really useful as a case for where hgvs may be getting it wrong and we are trying to do a better job.

[Peter-J-Freeman]

OK, and I have tested locally

http://127.0.0.1:8000/LOVD/lovd/GRCh37/%20NC_000022.10%3Ag.19929250_19929251insCCCCGCC/refseq/mane_select/False/primary?content-type=application%2Fjson

will now give

{
"NC_000022.10:g.19929250_19929251insCCCCGCC": {
"NC_000022.10:g.19929250_19929251insCCCCGCC": {
"g_hgvs": "NC_000022.10:g.19929252_19929258dup",
"genomic_variant_error": "NC_000022.10:g.19929250_19929251insCCCCGCC updated to NC_000022.10:g.19929252_19929258dup",
"genomic_variant_warnings": null,
"hgvs_t_and_p": {
"NM_006440.5": {
"alt_genomic_loci": [],
"gap_statement": null,
"gapped_alignment_warning": null,
"gene_info": {
"hgnc_id": "HGNC:18155",
"symbol": "TXNRD2"
},
"p_hgvs_slc": "NP_006431.2:p.V26Gfs*132",
"p_hgvs_tlc": "NP_006431.2:p.Val26GlyfsTer132",
"primary_assembly_loci": {
"grch37": {
"NC_000022.10": {
"hgvs_genomic_description": "NC_000022.10:g.19929252_19929258dup",
"vcf": {
"alt": "ACCCCGCC",
"chr": "22",
"pos": "19929250",
"ref": "A"
}
}
},
"grch38": {
"NC_000022.11": {
"hgvs_genomic_description": "NC_000022.11:g.19941729_19941735dup",
"vcf": {
"alt": "ACCCCGCC",
"chr": "22",
"pos": "19941727",
"ref": "A"
}
}
},
"hg19": {
"NC_000022.10": {
"hgvs_genomic_description": "NC_000022.10:g.19929252_19929258dup",
"vcf": {
"alt": "ACCCCGCC",
"chr": "chr22",
"pos": "19929250",
"ref": "A"
}
}
},
"hg38": {
"NC_000022.11": {
"hgvs_genomic_description": "NC_000022.11:g.19941729_19941735dup",
"vcf": {
"alt": "ACCCCGCC",
"chr": "chr22",
"pos": "19941727",
"ref": "A"
}
}
}
},
"select_status": {
"mane_select": true
},
"t_hgvs": "NM_006440.5:c.70_76dup",
"transcript_variant_error": null,
"transcript_version_warning": null
}
},
"p_vcf": "22:19929250:A:ACCCCGCC",
"selected_build": "GRCh37"
},
"errors": [],
"flag": "genomic_variant_warning"
},
"metadata": {
"variantformatter_version": "3.0.1.dev39+g8d54d3459",
"variantvalidator_hgvs_version": "2.2.1.dev22+gebef587af",
"variantvalidator_version": "3.0.2.dev221+g84abea0ed.d20260526",
"vvdb_version": "vvdb_2025_3",
"vvseqrepo_db": "VV_SR_2025_02/master",
"vvta_version": "vvta_2025_02"
}
}

so the bug you identified was correct, but we need to confirm the translation is correct.

FYI, {
"error": "NP_006431.2:p.(Val26_Lys158delinsGlyArgGlyAlaGlyArgGlyAlaGlyArgSerSerArgSerAlaGlyLeuTerSerProGlyGlyArgArgGlyIleTrpTrpProGlyLeuCysGlnGlyGlyArgProAlaGlyLysGluGlyGlyArgGlyGlyLeuArgGlyThrPheSerProArgHisProValGlyProArgArgHisLeuArgGlnArgGlyLeuHisProGlnGluAlaAspAlaProGlyGlyThrAlaGlyArgProAspProArgCysProGlnLeuTrpLeuGlyGlyGlyProAlaArgAlaAlaTerLeuGluGluAspGlyArgSerCysSerLysSerArgGluIleLeuGluLeuGlyProProCysProAlaSerGlyGlnLysSerGlnValLeuTer): char 87: expected the character ')'"
}
came from hgvs.py its a textbook error format :). I suspect that it mishandled the translation

[Peter-J-Freeman]

Mutalyzer gives
https://mutalyzer.nl/normalizer/NC_000022.10(NM_006440.5):c.70_76dup

Genomic Description
NC_000022.10:g.19929252_19929258dup
Coding Transcript Description
Gene: TXNRD2
NC_000022.10(NM_006440.5):c.70_76dup

Predicted RNA Description
NC_000022.10(NM_006440.5):r.(69_75delinsgggcggggggcggg)

Predicted Protein Description
NC_000022.10(NP_006431.2):p.(Val26GlyfsTer18)

We got NM_006440.5:c.70_76dup so yes, the translation could well be wrong still.
We have currently NP_006431.2:p.Val26GlyfsTer132 so the fs position is the same at Val26, but the ter is (posibly) way too long as observed.
Thanks for the bug. I will look deeper.

I a confused by Predicted RNA Description
[NC_000022.10(NM_006440.5):r.(69_75delinsgggcggggggcggg)]. the r. should match the c. in this case no? Something is really odd

[Peter-J-Freeman]

DNA level working

Reference sequence:
AGAAGCCCCACGACGATGGCGGCAATGGCGGTGGCGCTGCGGGGATTAGGAGGGCGCTTCCGGTGGCGGACGCAGGCCGTGGCGggcggggTGCGGGGCGCGGCGCGGGGCGCAGCAGCAGGTCAGCGGGACTATGATCTCCTGGTGGTCGGCGGGGGATCTGGTGGCCTGGCTTGTGCCAAGGAGGCCGCCCAGCTGGGAAGGAAGGTGGCCGTGGTGGACTACGTGGAACCTTCTCCCCAAGGCACCCGGTGGGGCCTCGGCGGCACCTGCGTCAACGTGGGCTGCATCCCCAAGAAGCTGATGCACCAGGCGGCACTGCTGGGAGGCCTGATCCAAGATGCCCCCAACTATGGCTGGGAGGTGGCCCAGCCCGTGCCGCATGACTGGAGGAAGATGGCAGAAGCTGTTCAAAATCACGTGAAATCCTTGAACTGGGGCCACCGTGTCCAGCTTCAGGACAGAAAAGTCAAGTACTTTAACATCAAAGCCAGCTTTGTTGACGAGCACACGGTTTGCGGCGTTGCCAAAGGTGGGAAAGAGATTCTGCTGTCAGCCGATCACATCATCATTGCTACTGGAGGGCGGCCGAGATACCCCACGCACATCGAAGGTGCCTTGGAATATGGAATCACAAGTGATGACATCTTCTGGCTGAAGGAATCCCCTGGAAAAACGTTGGTGGTCGGGGCCAGCTATGTGGCCCTGGAGTGTGCTGGCTTCCTCACCGGGATTGGGCTGGACACCACCATCATGATGCGCAGCATCCCCCTCCGCGGCTTCGACCAGCAAATGTCCTCCATGGTCATAGAGCACATGGCATCTCATGGCACCCGGTTCCTGAGGGGCTGTGCCCCCTCGCGGGTCAGGAGGCTCCCTGATGGCCAGCTGCAGGTCACCTGGGAGGACAGCACCACCGGCAAGGAGGACACGGGCACCTTTGACACCGTCCTGTGGGCCATAGGTCGAGTCCCAGACACCAGAAGTCTGAATTTGGAGAAGGCTGGGGTAGATACTAGCCCCGACACTCAGAAGATCCTGGTGGACTCCCGGGAAGCCACCTCTGTGCCCCACATCTACGCCATTGGTGACGTGGTGGAGGGGCGGCCTGAGCTGACACCCATAGCGATCATGGCCGGGAGGCTCCTGGTGCAGCGGCTCTTCGGCGGGTCCTCAGATCTGATGGACTACGACAATGTTCCCACGACCGTCTTCACCCCGCTGGAGTATGGCTGTGTGGGGCTGTCCGAGGAGGAGGCAGTGGCTCGCCACGGGCAGGAGCATGTTGAGGTCTATCACGCCCATTATAAACCACTGGAGTTCACGGTGGCTGGACGAGATGCATCCCAGTGTTATGTAAAGATGGTGTGCCTGAGGGAGCCCCCACAGCTGGTGCTGGGCCTGCATTTCCTTGGCCCCAACGCAGGCGAAGTTACTCAAGGATTTGCTCTGGGGATCAAGTGTGGGGCTTCCTATGCGCAGGTGATGCGGACCGTGGGTATCCATCCCACATGCTCTGAGGAGGTAGTCAAGCTGCGCATCTCCAAGCGCTCAGGCCTGGACCCCACGGTGACAGGCTGCTGAGGGTAAGCGCCATCCCTGCAGGCCAGGGCACACGGTGCGCCCGCCGCCAGCTCCTCGGAGGCCAGACCCAGGATGGCTGCAGGCCAGGTTTGGGGGGCCTCAACCCTCTCCTGGAGCGCCTGTGAGATGGTCAGCGTGGAGCGCAAGTGCTGGACAGGTGGCCCGTGTGCCCCACAGGGATGGCTCAGGGGACTGTCCACCTCACCCCTGCACCTCTCAGCCTCTGCCGCCGGGCACCCCCCCCCAGGCTCCTGGTGCCAGATGATGACGACCTGGGTGGAAACCTACCCTGTGGGCACCCATGTCCGAGCCCCCTGGCATTTCTGCAATGCAAATAAAGAGGGTACTTTTTCTGAA
Deletion sequence:
GGCGGGG
Inserted sequence:
GGCGGGGGGCGGGG
Var sequence:
AGAAGCCCCACGACGATGGCGGCAATGGCGGTGGCGCTGCGGGGATTAGGAGGGCGCTTCCGGTGGCGGACGCAGGCCGTGGCGGGCGGGGGGCGGGGTGCGGGGCGCGGCGCGGGGCGCAGCAGCAGGTCAGCGGGACTATGATCTCCTGGTGGTCGGCGGGGGATCTGGTGGCCTGGCTTGTGCCAAGGAGGCCGCCCAGCTGGGAAGGAAGGTGGCCGTGGTGGACTACGTGGAACCTTCTCCCCAAGGCACCCGGTGGGGCCTCGGCGGCACCTGCGTCAACGTGGGCTGCATCCCCAAGAAGCTGATGCACCAGGCGGCACTGCTGGGAGGCCTGATCCAAGATGCCCCCAACTATGGCTGGGAGGTGGCCCAGCCCGTGCCGCATGACTGGAGGAAGATGGCAGAAGCTGTTCAAAATCACGTGAAATCCTTGAACTGGGGCCACCGTGTCCAGCTTCAGGACAGAAAAGTCAAGTACTTTAACATCAAAGCCAGCTTTGTTGACGAGCACACGGTTTGCGGCGTTGCCAAAGGTGGGAAAGAGATTCTGCTGTCAGCCGATCACATCATCATTGCTACTGGAGGGCGGCCGAGATACCCCACGCACATCGAAGGTGCCTTGGAATATGGAATCACAAGTGATGACATCTTCTGGCTGAAGGAATCCCCTGGAAAAACGTTGGTGGTCGGGGCCAGCTATGTGGCCCTGGAGTGTGCTGGCTTCCTCACCGGGATTGGGCTGGACACCACCATCATGATGCGCAGCATCCCCCTCCGCGGCTTCGACCAGCAAATGTCCTCCATGGTCATAGAGCACATGGCATCTCATGGCACCCGGTTCCTGAGGGGCTGTGCCCCCTCGCGGGTCAGGAGGCTCCCTGATGGCCAGCTGCAGGTCACCTGGGAGGACAGCACCACCGGCAAGGAGGACACGGGCACCTTTGACACCGTCCTGTGGGCCATAGGTCGAGTCCCAGACACCAGAAGTCTGAATTTGGAGAAGGCTGGGGTAGATACTAGCCCCGACACTCAGAAGATCCTGGTGGACTCCCGGGAAGCCACCTCTGTGCCCCACATCTACGCCATTGGTGACGTGGTGGAGGGGCGGCCTGAGCTGACACCCATAGCGATCATGGCCGGGAGGCTCCTGGTGCAGCGGCTCTTCGGCGGGTCCTCAGATCTGATGGACTACGACAATGTTCCCACGACCGTCTTCACCCCGCTGGAGTATGGCTGTGTGGGGCTGTCCGAGGAGGAGGCAGTGGCTCGCCACGGGCAGGAGCATGTTGAGGTCTATCACGCCCATTATAAACCACTGGAGTTCACGGTGGCTGGACGAGATGCATCCCAGTGTTATGTAAAGATGGTGTGCCTGAGGGAGCCCCCACAGCTGGTGCTGGGCCTGCATTTCCTTGGCCCCAACGCAGGCGAAGTTACTCAAGGATTTGCTCTGGGGATCAAGTGTGGGGCTTCCTATGCGCAGGTGATGCGGACCGTGGGTATCCATCCCACATGCTCTGAGGAGGTAGTCAAGCTGCGCATCTCCAAGCGCTCAGGCCTGGACCCCACGGTGACAGGCTGCTGAGGGTAAGCGCCATCCCTGCAGGCCAGGGCACACGGTGCGCCCGCCGCCAGCTCCTCGGAGGCCAGACCCAGGATGGCTGCAGGCCAGGTTTGGGGGGCCTCAACCCTCTCCTGGAGCGCCTGTGAGATGGTCAGCGTGGAGCGCAAGTGCTGGACAGGTGGCCCGTGTGCCCCACAGGGATGGCTCAGGGGACTGTCCACCTCACCCCTGCACCTCTCAGCCTCTGCCGCCGGGCACCCCCCCCCAGGCTCCTGGTGCCAGATGATGACGACCTGGGTGGAAACCTACCCTGTGGGCACCCATGTCCGAGCCCCCTGGCATTTCTGCAATGCAAATAAAGAGGGTACTTTTTCTGAA

So this looks OK so far. I lower cases the reference sequence to help locate the edit

[ifokkema]

I've taken the NM, applied the duplication, and I get:
atggcggcaatggcggtggcgctgcggggattaggagggcgcttccggtggcggacgcaggccgtggcgggcggggggcggggtgcggggcgcggcgcggggcgcagcagcaggtcagcgggactatga(etc)
which then translates into MAAMAVALRGLGGRFRWRTQAVAGGGRGAGRGAGRSSRSAGL*, which then, compared to the reference AA code from the NM, would be p.(V26Gfs18). So Mutalyzer's correct on this one.

[ifokkema]

For reference. The HGVS Python library was previously used for insertion translations. It keeps messing up, so I have now passed it into our own translating code. This is really useful as a case for where hgvs may be getting it wrong and we are trying to do a better job.

Interesting; I would suspect that translating a generated sequence is rather straightforward, but it does indeed seem that they messed up somewhere.

[ifokkema]

I a confused by Predicted RNA Description [NC_000022.10(NM_006440.5):r.(69_75delinsgggcggggggcggg)]. the r. should match the c. in this case no? Something is really odd

Euh, yeah, that's not normal... why would the RNA prediction be different? 🤔

[Peter-J-Freeman]

I seem to be getting the same mRNA change, can you check the rest of the sequence to double check my working. Then I will look at the translation and why it failed to terminate. Maybe it accidentally took the Selenon dict :)

Var sequence:
AGAAGCCCCACGACGATGGCGGCAATGGCGGTGGCGCTGCGGGGATTAGGAGGGCGCTTCCGGTGGCGGACGCAGGCCGTGGCGGGCGGGGGGCGGGGTGCGGGGCGCGGCGCGGGGCGCAGCAGCAGGTCAGCGGGACTATGATCTCCTGGTGGTCGGCGGGGGATCTGGTGGCCTGGCTTGTGCCAAGGAGGCCGCCCAGCTGGGAAGGAAGGTGGCCGTGGTGGACTACGTGGAACCTTCTCCCCAAGGCACCCGGTGGGGCCTCGGCGGCACCTGCGTCAACGTGGGCTGCATCCCCAAGAAGCTGATGCACCAGGCGGCACTGCTGGGAGGCCTGATCCAAGATGCCCCCAACTATGGCTGGGAGGTGGCCCAGCCCGTGCCGCATGACTGGAGGAAGATGGCAGAAGCTGTTCAAAATCACGTGAAATCCTTGAACTGGGGCCACCGTGTCCAGCTTCAGGACAGAAAAGTCAAGTACTTTAACATCAAAGCCAGCTTTGTTGACGAGCACACGGTTTGCGGCGTTGCCAAAGGTGGGAAAGAGATTCTGCTGTCAGCCGATCACATCATCATTGCTACTGGAGGGCGGCCGAGATACCCCACGCACATCGAAGGTGCCTTGGAATATGGAATCACAAGTGATGACATCTTCTGGCTGAAGGAATCCCCTGGAAAAACGTTGGTGGTCGGGGCCAGCTATGTGGCCCTGGAGTGTGCTGGCTTCCTCACCGGGATTGGGCTGGACACCACCATCATGATGCGCAGCATCCCCCTCCGCGGCTTCGACCAGCAAATGTCCTCCATGGTCATAGAGCACATGGCATCTCATGGCACCCGGTTCCTGAGGGGCTGTGCCCCCTCGCGGGTCAGGAGGCTCCCTGATGGCCAGCTGCAGGTCACCTGGGAGGACAGCACCACCGGCAAGGAGGACACGGGCACCTTTGACACCGTCCTGTGGGCCATAGGTCGAGTCCCAGACACCAGAAGTCTGAATTTGGAGAAGGCTGGGGTAGATACTAGCCCCGACACTCAGAAGATCCTGGTGGACTCCCGGGAAGCCACCTCTGTGCCCCACATCTACGCCATTGGTGACGTGGTGGAGGGGCGGCCTGAGCTGACACCCATAGCGATCATGGCCGGGAGGCTCCTGGTGCAGCGGCTCTTCGGCGGGTCCTCAGATCTGATGGACTACGACAATGTTCCCACGACCGTCTTCACCCCGCTGGAGTATGGCTGTGTGGGGCTGTCCGAGGAGGAGGCAGTGGCTCGCCACGGGCAGGAGCATGTTGAGGTCTATCACGCCCATTATAAACCACTGGAGTTCACGGTGGCTGGACGAGATGCATCCCAGTGTTATGTAAAGATGGTGTGCCTGAGGGAGCCCCCACAGCTGGTGCTGGGCCTGCATTTCCTTGGCCCCAACGCAGGCGAAGTTACTCAAGGATTTGCTCTGGGGATCAAGTGTGGGGCTTCCTATGCGCAGGTGATGCGGACCGTGGGTATCCATCCCACATGCTCTGAGGAGGTAGTCAAGCTGCGCATCTCCAAGCGCTCAGGCCTGGACCCCACGGTGACAGGCTGCTGAGGGTAAGCGCCATCCCTGCAGGCCAGGGCACACGGTGCGCCCGCCGCCAGCTCCTCGGAGGCCAGACCCAGGATGGCTGCAGGCCAGGTTTGGGGGGCCTCAACCCTCTCCTGGAGCGCCTGTGAGATGGTCAGCGTGGAGCGCAAGTGCTGGACAGGTGGCCCGTGTGCCCCACAGGGATGGCTCAGGGGACTGTCCACCTCACCCCTGCACCTCTCAGCCTCTGCCGCCGGGCACCCCCCCCCAGGCTCCTGGTGCCAGATGATGACGACCTGGGTGGAAACCTACCCTGTGGGCACCCATGTCCGAGCCCCCTGGCATTTCTGCAATGCAAATAAAGAGGGTACTTTTTCTGAA

[ifokkema]

I seem to be getting the same mRNA change, can you check the rest of the sequence to double check my working. Then I will look at the translation and why it failed to terminate. Maybe it accidentally took the Selenon dict :)

You will need to remove the first 15 nucleotides to get the CDS, but yeah, that's the variant sequence.

[Peter-J-Freeman]

@ifokkema The reference protein has Sec in it

RefSeq: MAAMAVALRGLGGRFRWRTQAVAGGVRGAARGAAAGQRDYDLLVVGGGSGGLACAKEAAQLGRKVAVVDYVEPSPQGTRWGLGGTCVNVGCIPKKLMHQAALLGGLIQDAPNYGWEVAQPVPHDWRKMAEAVQNHVKSLNWGHRVQLQDRKVKYFNIKASFVDEHTVCGVAKGGKEILLSADHIIIATGGRPRYPTHIEGALEYGITSDDIFWLKESPGKTLVVGASYVALECAGFLTGIGLDTTIMMRSIPLRGFDQQMSSMVIEHMASHGTRFLRGCAPSRVRRLPDGQLQVTWEDSTTGKEDTGTFDTVLWAIGRVPDTRSLNLEKAGVDTSPDTQKILVDSREATSVPHIYAIGDVVEGRPELTPIAIMAGRLLVQRLFGGSSDLMDYDNVPTTVFTPLEYGCVGLSEEEAVARHGQEHVEVYHAHYKPLEFTVAGRDASQCYVKMVCLREPPQLVLGLHFLGPNAGEVTQGFALGIKCGASYAQVMRTVGIHPTCSEEVVKLRISKRSGLDPTVTGCUG*

Second to last amino acid.

VV is correctly, all be it naievely, applying Ter > Sec for the correct Termination. look at the related issue. Mutalyzer is ignoring this!
Confirms its a selenoprotein https://www.ncbi.nlm.nih.gov/nuccore/NM_006440.5/