Downstream analysis tools pVACbind

[martinez14recerca]

Hello,
I have been using pVACbind to select candidate neoantigens for class I and II. For every single patient, I have worked with the "all.epitopes.tsv" using the Median IC50 Score to determine weak binders and strong binders. From this results, I have generated a second .tsv containing the following information: Chromosome, Start, Stop, Reference, Variant, Transcript, HLA Allele, Peptide length, Epitope Sequence (as shown in the document attached. Part of the information is given by pVACbind and the rest is from previous analysis that I have). I have generated single .tsv for each patient (without adding the patient and WB_SB columns).

Now my intention is to run the downstream analysis tools (NetChop predict cleavages and NetMHCStab predict stability). Since I am not using the direct "all.epitopes.tsv", I run the following command in my computer:
pvacbind run \ /home/immunocart/ANALISISTCRDOWNSTREAM/LLC_001_I.tsv \ LLC_001_I \ HLA-A*11:01,HLA-A*24:02,HLA-C*16:01 \ all \ /home/immunocart/ANALISISTCRDOWNSTREAM/RESULTS/LLC_001_I_output \ --net-chop-method cterm \ --netmhc-stab \ -e1 8,9,10,11
It seems the code is running but at the end I am not getting any file:
Executing MHC Class I predictions Splitting FASTA into smaller chunks Splitting FASTA into smaller chunks - Entries 1-2 Completed Fasta file /home/immunocart/ANALISISTCRDOWNSTREAM/RESULTS/LLC_001_I_output/MHC_Class_I/tmp/LLC_001_I.8.fa.split_1-2 is empty. Skipping Splitting FASTA into smaller chunks Splitting FASTA into smaller chunks - Entries 1-2 Completed Fasta file /home/immunocart/ANALISISTCRDOWNSTREAM/RESULTS/LLC_001_I_output/MHC_Class_I/tmp/LLC_001_I.9.fa.split_1-2 is empty. Skipping Splitting FASTA into smaller chunks Splitting FASTA into smaller chunks - Entries 1-2 Completed Fasta file /home/immunocart/ANALISISTCRDOWNSTREAM/RESULTS/LLC_001_I_output/MHC_Class_I/tmp/LLC_001_I.10.fa.split_1-2 is empty. Skipping Splitting FASTA into smaller chunks Splitting FASTA into smaller chunks - Entries 1-2 Completed Fasta file /home/immunocart/ANALISISTCRDOWNSTREAM/RESULTS/LLC_001_I_output/MHC_Class_I/tmp/LLC_001_I.11.fa.split_1-2 is empty. Skipping No output files were created. Aborting. No MHC class II alleles chosen. Skipping MHC class II predictions.

I would like to ask if you could detect the error or propose another alternative, since I thought the input file contains all the necessary information for the programs to run. Thanks in advance!

GitHub_20241031.xlsx

[susannasiebert]

I don't think rerunning pvacbind run is the correct approach here. pVACbind expects a FASTA file, not a TSV. Please have a look at the optional downstream analysis tools section of the documentation particularly the two commands to run NetChop and NetMHCstabpan standalone. As long as your TSV file contains the Mutation and Epitope Seq columns, this should work. These two columns correspond to the FASTA header of the original input fasta entry the epitope came from as well as the epitope, respectively.