Annotate VCF files - soprano-app

[faizjab]

Hi,

I am trying to run step 2 on soprano-app to annotate my VCF file. However, I get this error:

As I am trying to create the vcf.anno file through step 2, I am unsure how I am supposed to have the file already created? Any help would be much appreciated.

[faizjab]

I then created an empty test.vcf.anno file to get around this and annotation was succesfull. But it created an empty file.

[faizjab]

@rachelicr

[rachelicr]

@faizjab if this is a support request from the ICR could you mail schelpdesk@icr.ac.uk so that we can filter it though to the correct place?

[rachelicr]

@faizjab This error is because you haven't given a name in the box where it says "Choose a name for the annotated output". This is not optional - this should be clearer, and should default to something temporary.

[rachelicr]

I will add making this clearer as something in our todo list,but meanwhile you should be able to proceed by simply giving a name.

[faizjab]

@rachelicr. Thank you for the reply. Even when putting in a name in the box for "Choose a name for the annotated output" I still get the same error.

However I ended up using the VEP tool on the Ensembl website to get around this.

The error I get now when running SOPRANO is this:

[rachelicr]

If you attach the data you are using I can have a look?

[faizjab]

tcga_full_selected.annotated.zip

This is a VEP of TCGA LUSC 2018.

[rachelicr]

What about the original vcf you say fails in the first step? test.vcf

[faizjab]

lusc_tcga_2018.vcf.zip Here's the original vcf file for TCGA LUSC 2018, thank you for helping with this

[rachelicr]

The problem with the original file is that the annotator expects some header data, specifically in complains:

Error in vcfR::read.vcfR(vcf_path, verbose = FALSE) : 
  File: /home/ralcraft/dev/shiny-proxy-developed/SOPRANO/data/test_data/faizjab_jun24/lusc_tacg_2018.vcf does not appear to be a VCF file.
  First line of file:
 /home/ralcraft/dev/shiny-proxy-developed/SOPRANO/data/test_data/faizjab_jun24/lusc_tacg_2018.vcf 
  Should begin with:
##fileformat=VCFv 

A file that works has the first 2 rows as:

fileformat=VCFv4.1

fileDate=20240527

[faizjab]

Hi,

Thank you for this.

Even when I add the first two rows as above, I get the same error.

When using Ensembl's VEP annotation function I am able to annotate the VCF file and run SOPRANO.

However the error I get now says this. I am using the allhlabinders_exprmean1.IEDBpeps.bed file

[rachelicr]

It needs to be in fully specified vcf format: https://samtools.github.io/hts-specs/VCFv4.2.pdf

I suspect the next step isn;t going to work if the original file doesn;t work so it is this step that needs to be fixed.

[faizjab]

I have formatted the data into a fully specified vcf format but still get the same error at step 2

[rachelicr]

Can attach the fully specified file?

[faizjab]

output.vcf.zip

Here's the vcf file

[rachelicr]

That file has worked for me both from the command line and the web-app:

[faizjab]

thank you!

[faizjab]

@rachelicr when using the soprano web app I am trying to get the GRCh37 reference genome. I have this already downloaded on my computer but I can't seem to link to it. Would you be able to advise?

[rachelicr]

If you are using the application, you should be able to select GRCh37 form the drop down?

The web application has this functionality restricted to only those we have predownloaded to reduce space. If you want to link to your own genome cache then you need to run the application locally either in docker or from the clone repo, or from the command line.

https://instituteofcancerresearch.github.io/SOPRANO/installation/