VCF2MAF structural variant file issue.

Question

[The_PyPanda]

Hello,

I am trying to convert VCF files for tumour samples into MAF files. To perform MAFtools analysis. I have succeeded in doing this for somatic variants, but am struggling to convert VCF files for structural variants generated using Manta.

Screenshots of the error output I get are:

[faidx] Failed to fetch sequence in : - 
Use of uninitialized value in join or string at /../vac2maf.pl line 291 <GENO> line 2638

After repetition of the above, my output ends with the following:

Argument “” isn’t numeric in subtraction (-) at  /../vac2maf.pl line 291 <GENO> line 2638
[W::fai_get_val] Reference : -1-1 not found in FASTA file, returning empty sequence. 
[faidx] Failed to fetch sequence in: -1-1
ERROR: : You’re either using an outdated samtools, or –ref-fas is not the sae genome build as your –input-vcf

(Apologies – I can’t seem to copy screenshot in the question box).

I am using the same reference genome/index file that I used to call my variants. Any suggestions/advice would be greatly appreciated.

Thank You

[Ram]

Argument “” isn’t numeric in subtraction (-) at /../vac2maf.pl line 291 <GENO> line 2638

These stylized quotes should not be in the script - it means the script was formatted using a word processor such as Microsoft Word at some point. Where did you get the scripts from?

[The_PyPanda]

Hi the following is my script. I can't copy it out of server and can't seem to post screen shots here:

module load /VEP/99.1-foss-2019a-Perl-5.28.1

perl /vcf2maf/1.6.17/vcf2maf.pl \
--input-vcf \
--output \
--tumour-id \
--normal-id Germline \
--ref-fasta no_alt.fa \
--vep-path \
--vep-data \
--filter-vcf 0 \
--ncbi-build GRCh38 \
--cache-version 99

I can't see any stylized quotes. Do I need to change a command function? Or do you suggest an alternative script?

[Ram]

Please use Add Comment/Add Reply. Answers are reserved for posts that answer the top-level question.

Are you sure module load /VEP... is correct? Check with your sysadmin - no module load command I've ever used took arguments starting with a /, as the command looks in a modulefiles directory for a subdirectory matching the module name.

Also, you're loading the module the invoking a script from an odd location (/vcf2maf/). Either your sysadmin has designed software program to be mounted as drives (a really odd architecture) or you're using things wrong.

As to using the forum, use the 101010 button to format code (I've done it for you this time) and use the image button (the one to the right of the 101010 button to upload a file from your computer)

Here's a tutorial I wrote on formatting code: How to Use Biostars Part-3: Formatting Text and Using GitHub Gists

[The_PyPanda]

Sorry to keep it simple I didn't type out the full path.

But in relation to your above comments. My full paths for:

(A) Module load

module load bio/VEP/99.1-foss-2019a-Perl-5.28.1

(B) perl

perl /resources/tools/manual_apps/software/bio/vcf2maf//1.6.17/vcf2maf.pl \

Of a secondary note. I attempted to run the command

--inhibit-VEP

I inserted this command prior to input-vcf. But this did not work. I got the error:

'Unknown option: inhibit-vep'

Any suggestions?

[GenoMax]

Is all of this in a file that you are submitting the scheduler or are you running directly on the command line?

[The_PyPanda]

Yes above is the full script. Submitting to the HPC server.

[Ram]

What is the samtools version on the server? The error message tells you that your samtools could be outdated. Also, is your reference fasta the same version used to generate the VCF file? These are the two issues you need to cover before going further into debugging.

[The_PyPanda]

Hi Ram, I have looked into these, as my first step. I am using the same Fasta file I did to generate my vcf file and samtools release 1.18. But unfortunately no look. It works fine for the somatic variants though.

[Ram]

Are you sure you're copy-pasting your error(s) properly (copying directly from the terminal and pasting here with no word processing software in between and also not typing the error)? There is a typo in your error message that cannot exist - that typo has never existed in the history of vcf2maf.pl.

[The_PyPanda]

I have taken a screenshots of the error that I receive for clarity. ![enter image description here][1]

[Ram]

OK that clarifies two things - you did not have stylized quotes in the error message, and you don't have that typo ("same" misspelled as "sae"). These copy-paste errors take away from our ability to figure out the actual problem, so please be more careful in the future.

Can you show me the output to bcftools view -H your_input.vcf | head | cut -f 1-5?

[The_PyPanda]

I have attached a a screen shot of my input vcf. Unfortunately its not generating an output vcf yet.

[Ram]

If things go well, your output will be MAF, not VCF. The VCF looks fine. Do you have a bioinformatician near you that you can consult? This is not going to be even a tiny bit easy to debug over the Internet.

[Ram]

Once again, you added a comment as an answer. Please stop doing this. If you're unable to figure out how to add a comment-reply, see this: How to Use Biostars, Part-I: Questions, Answers, Comments and Replies