Another Noobie question

[WalterPavicic]

Hi, I'm starting to figured out how to run sigflow. I managed to run test files.
However, when trying to run my own .vcf file I get the following error message:

User command calling: /usr/lib/R/bin/exec/R --no-echo --no-restore --file=/usr/local/bin/sigflow --args extract -i /home/walter/Documents/Mutational_Signature/varscan_tumor_normal.snp_somatic_calls.annot_significant.vcf -o /home/walter/Sigflow/sigflow/test_results/test_046 -m MAF -r 10 -T 4 --max 10

Reading file...

Try reading as a MAF file.
No 'Tumor_Sample_Barcode' column detected, try parsing as a component-by-sample matrix.
Error in .rowNamesDF<-(x, value = value) :
duplicate 'row.names' are not allowed
Calls: rownames<- ... row.names<- -> row.names<-.data.frame -> .rowNamesDF<-
In addition: Warning message:
non-unique values when setting 'row.names': ‘chr1’, ‘chr10’, ‘chr11’, ‘chr12’, ‘chr13’, ‘chr14’, ‘chr15’, ‘chr16’, ‘chr17’, ‘chr18’, ‘chr19’, ‘chr2’, ‘chr20’, ‘chr21’, ‘chr22’, ‘chr3’, ‘chr4’, ‘chr5’, ‘chr6’, ‘chr7’, ‘chr8’, ‘chr9’, ‘chrX’
Execution halted

In brief, my .vcf was generated by using the next pipeline:
.bam files (N and T), samtools -> .pileup files
.pileup -> varscan2 -> N-T.vcf
N-T.vcf -> Data Annotation with SNPeff/snpsift -> N-T.annot.vcf
N-T.annot.vcf -> filtering VCF to pick up <0.01 variants (using 'VariantAnnotation') -> N-T.annot.signif.vcf, this .vcf file was used as SigFlow input.

Hoping you can give me any insight on were is the mistake on my .vcf

Regards.

[ShixiangWang]

Hi could you try set -m to SBS and re-run? or unset this option?

-m <class>, --mode <class>      extract/fit mode, can be one of SBS, DBS, ID, MAF (for three types), CN (not supported in fit subcommand) [default: SBS].

[ShixiangWang]

In test directory, I provide test datasets and script https://github.com/ShixiangWang/sigflow/blob/master/test/test.sh. For vcfs, you should put them in a directory and specify the -i option to the directory path.

https://github.com/ShixiangWang/sigflow/blob/a7e2ae78af8f0d4ef354a027b92cd177afbceefd/test/test.sh#L31

[WalterPavicic]

Hi, thanks for your reply.
if I got it right...I follow the L31 script: sigflow extract -i vcf/ -o test_results/test_vcf -m MAF -r 10 -T 4 --max 10
I put my .vcf file in the VCF directory inside the SigFlow folders, modifying the -i and -o parameters as this:
~/Sigflow/sigflow$ sigflow extract -i test/vcf/varscan_tumor_normal.snp_somatic_calls.annot_significant.vcf -o test/test_results/Test_046 -m MAF -r 10 -T 4 --max 10

again the same error.
So, not sure if Im not doing what your suggested or my VCF file is corrupted and/or not in the required format.

My VCF file looks like this (which is not the same look as your's, could it be the reason?:

#CHROM POS ID REF ALT QUAL FILTER INFO FORMAT NORMAL TUMOR
chr1 1115664 chr1:1115664_C/A C A . PASS DP=234;SOMATIC;SS=2;SSC=35;GPV=1;SPV=0.00026878;EFF=synonymous_variant(LOW|SILENT|ccC/ccA|p.Pro150Pro/c.450C>A|673|TTLL10|protein_coding|CODING|ENST00000379289|6|A) GT:DP:RD:AD:FREQ:DP4 0/0:101:101:0:0%:46 0/1:133:119:14:10.53%:46

[ShixiangWang]

For vcfs, you should put them in a directory and specify the -i option to the directory path.

[WalterPavicic]

Hi!
Finally, I managed to run SigFlow analysis using my .vcf files!
I'm new on this, that's why it took me a while...

so, once inside the right directory, the command was:
(base) walter@Helsinki:~/Sigflow/sigflow/test$ sigflow extract -i vcf/ -o test_results/test_vcf -m MAF -r 10 -T 4 --max 10

thanks for your support! and a great tool!

[ShixiangWang]

you are welcome :).