pVACseq: Discrepancy between {sample_id}.all_epitopes.tsv and {sample_id}.fasta ?

[yiolino]

Hi, thank you for developing such an excellent software.

pVACseq: Discrepancy between {sample_id}.all_epitopes.tsv and {sample_id}.fasta ?

Description

I have successfully run pVACseq without any errors.
However, I have observed that certain peptides listed in the {sample_id}.all_epitopes.tsv are not present in the {sample_id}.fasta .

Expected Behavior

I expected that all peptides listed in the {sample_id}.all_epitopes.tsv would be included in the {sample_id}.fasta.

Actual Behavior

Certain peptides present in the {sample_id}.all_epitopes.tsv are missing from the {sample_id}.fasta.

Questions

1. Is it expected behavior that not all peptides listed in the {sample_id}.all_epitopes.tsv file are included in the {sample_id}.fasta?
2. If this is expected, could you please explain the logic behind this behavior? I am particularly interested in the section of the code in generate_protein_fasta.py here. Is my assumption correct that this part of the code is related to the observed behavior?

Additional Information

• pVACseq version: Docker 3.0.1
• OS Ubuntu 22.04

[susannasiebert]

This is generally not expected. Would you be able to share your input VCF with us for further investigation? A VCF file with just the variant in question would be sufficient for us to be able to investigate this on our end.

[yiolino]

@susannasiebert
Sorry for the late reply.
I have created a VCF with unnecessary information masked.
I have sent it to you via email.
Please check it.

[susannasiebert]

I tried running pVACtools with the VCF you provided but I'm unable to replicate this issue. Can you please provide me with an example pVACseq command where you noted this problem?

[susannasiebert]

@yiolino I wanted to ping you about this issue. I was unable to replicate this problem so it would help if you could provide an example pVACseq command where you noticed this discrepancy.

[yiolino]

@susannasiebert
I apologize for the late reply.

The command I used is as follows:

pvacseq run \
    “$in_vcf” \
    “$sample_name” \
    “$HLAtype” \
    $binding_alogs \
    “$out_dir” \
    --n-threads “$threads” \
    --iedb-install-directory /opt/iedb \
    --trna-cov 0 \
    --trna-vaf 0 \
    --expn-val 0 \
    --phased-proximal-variants-vcf “$phased_vcf” \
    --pass-only

Would you need the contents of the phased VCF file to help replicate the issue?

Best regards,
yiolino

[susannasiebert]

I think I do, yes. Please also include the values of $HALtype and $binding_alogs for this particular run.

Additionally, if you could also attach the all_epitopes.tsv and fasta file where you noticed the discrepancy, that would be helpful in debugging this further.