Question: Error while running Varscan copynumber command
gravatar for manojkumar_bhosale
5.2 years ago by
manojkumar_bhosale70 wrote:


Q1) I am using Varscan for detecting the CNV and LOH variants from matched normal sample. I have generated two pileup files from sample BAM files using samtools. Below is the command i am using for generating the copynumber file. When I executed below command it ran for a while and generated copynumber file(~2 Mb's) and thrown and exception shown below.

    I would really appreciate if sombody tells me what mistake I have made here.

>java -jar VarScan.v2.3.7.jar copynumber normal.pileup tumor.pileup output.basename
Normal Pileup: normal.pileup
Tumor Pileup: tumor.pileup
Min coverage:   10
Min avg qual:   15
P-value thresh: 0.01
Exception in thread "main" java.lang.ArrayIndexOutOfBoundsException: 5
        at net.sf.varscan.Copynumber.<init>(
        at net.sf.varscan.VarScan.copynumber(
        at net.sf.varscan.VarScan.main(

Q2) When I use same files with combined mpileup below parsing exception is thrown

>java -jar VarScan.v2.3.7.jar copynumber normal_tumor.pileup testTN --mpileup 1

Min coverage:   10
Min avg qual:   15
P-value thresh: 0.01
Reading input from normal_tumor.pileup
Reading mpileup input...
Parsing Exception on line:
chr1    20141883        A       1       .$      >       0


Please help me to overcome these



ADD COMMENTlink modified 5.2 years ago • written 5.2 years ago by manojkumar_bhosale70

how did you generate the mpileup files ?

ADD REPLYlink written 5.2 years ago by Pierre Lindenbaum125k

Thanks for prompt reply !!!

Here are the commands I used to generate mpileup files

$./samtools mpileup -q 1 -f hg19.fasta normal.bam > normal.pileup

$./samtools mpileup -q 1 -f hg19.fasta tumor.bam > tumor.pileup

$./samtools mpileup -q 1 -l covered.bed -f hg19.fasta normal.bam tumor.bam > normal_tumor.pileup

P.S: All of the above mpileup files were generated without any error

ADD REPLYlink written 5.2 years ago by manojkumar_bhosale70

please see Problem generating Varscn2 copynumber output , which this question duplicates

You've passed in a line with no coverage in the right-most sample, you should filter these out before passing to VarScan

ADD REPLYlink modified 5.2 years ago • written 5.2 years ago by russhh5.0k

Thanks a lot !! The program runs fine now.

But, will the removal of loci with zero covarage add some bias to Copy number results? 

ADD REPLYlink written 5.2 years ago by manojkumar_bhosale70

Not at that line. You've implicitly told Varscan to only consider sites with a min coverage of 10, so it should disregard that site since the total coverage is 1.

Unless your tumour samples are completely devoid of fibroblasts, immune cells etc you should have some coverage at any site that has decent depth in your normal sample, so I wouldn't worry about this as a source of bias.

Identifying large regions that have zero coverage in your tumour and good depth in your normal sample should be really simple if you want to double check though.

ADD REPLYlink written 5.2 years ago by russhh5.0k

Could you please elaborate on what you mean by "a line with no coverage"? As per my understanding, the 4th column gives the number of reads covering a particular site. I changed all rows that had a 0 in column 4 to a 1 but that did not solve my problem. I am still getting the same error.

Thanks in advance for your help :)

ADD REPLYlink modified 16 days ago by RamRS25k • written 4.0 years ago by aditisk0
Please log in to add an answer.


Use of this site constitutes acceptance of our User Agreement and Privacy Policy.
Powered by Biostar version 2.3.0
Traffic: 1995 users visited in the last hour