VarScan early halt

allnovels · 01-20-2013, 10:40 AM

Hi,

I'm new to VarScan. I was trying to call copynumber from my exome sequencing data.

My script is
samtools mpileup -q 1 -f ~/hg19.fa normal.rmdup.bam tumor.rmdup.bam | java -Xmx4g -jar ~/VarScan.v2.3.3.jar copynumber > output --mpileup 1

It seems to run the first bit of the program and start to generate 2 files
1. copynumber
But it halted when it went to
chrM 16066 16165 100 53.7 89.7 0.740 1.0

The other file 30.copynumber only has the column name
chrom chr_start chr_stop num_positions normal_depth tumor_depth log2_ratio gc_content

Running message is as following:
[mpileup] 1 samples in 2 input files
<mpileup> Set max per-file depth to 8000
Min coverage: 10
Min avg qual: 15
P-value thresh: 0.01
Reading input from STDIN
Reading mpileup input...
Parsing Exception on line:
chrM 16184 C 0 0
8

It seems that something happened at the site chrM 16184... But I used my bam files to call SNPs/coverage from gatk and they worked out fine... I didn't know what's wrong...

Many thanks,
John

dkoboldt · 01-30-2013, 08:37 AM

John,

I would restructure your command slightly so that redirecting the output is the last thing you do... VarScan might not be interpreting your "--mpileup 1" :
samtools mpileup -q 1 -f ~/hg19.fa normal.rmdup.bam tumor.rmdup.bam | java -Xmx4g -jar ~/VarScan.v2.3.3.jar copynumber --mpileup 1 > output

If this doesn't help, I believe the issue occurs at a position where you have 0x coverage... those usually aren't present in pileup files. Please send me the mpileup lines for chrM from positions 16180- 16190 and I'll take a look.

Feel free to post here or e-mail to dkoboldt (at) genome [dot] wustl [dot] edu

lovenlong · 12-23-2013, 10:01 PM

Hi, all !

Though this thread had been posted about one year ago, I'd like to share my recent experience in meeting the same error in using Varscan2.
The causal thing would be the inconsistent of samtools version in generating BAMs and producing mpileup files.
In my case, input BAMs were generated by samtools-0.1.18 in weeks ago, while now the samtools version in my PATH is 0.1.19, and the error arised if I don't specify to use the samtools-0.1.18 in generating pileup file.

Thanks a bunch!

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Early experiences with 316 chips	nickloman	Ion Torrent	15	06-25-2012 08:17 PM

01-20-2013, 10:40 AM	#1
allnovels Junior Member Location: Los Angeles Join Date: Jan 2013 Posts: 1	VarScan early halt Hi, I'm new to VarScan. I was trying to call copynumber from my exome sequencing data. My script is samtools mpileup -q 1 -f ~/hg19.fa normal.rmdup.bam tumor.rmdup.bam \| java -Xmx4g -jar ~/VarScan.v2.3.3.jar copynumber > output --mpileup 1 It seems to run the first bit of the program and start to generate 2 files 1. copynumber But it halted when it went to chrM 16066 16165 100 53.7 89.7 0.740 1.0 The other file 30.copynumber only has the column name chrom chr_start chr_stop num_positions normal_depth tumor_depth log2_ratio gc_content Running message is as following: [mpileup] 1 samples in 2 input files <mpileup> Set max per-file depth to 8000 Min coverage: 10 Min avg qual: 15 P-value thresh: 0.01 Reading input from STDIN Reading mpileup input... Parsing Exception on line: chrM 16184 C 0 0 8 It seems that something happened at the site chrM 16184... But I used my bam files to call SNPs/coverage from gatk and they worked out fine... I didn't know what's wrong... Many thanks, John

01-30-2013, 08:37 AM	#2
dkoboldt Member Location: St. Louis Join Date: Mar 2009 Posts: 62	John, I would restructure your command slightly so that redirecting the output is the last thing you do... VarScan might not be interpreting your "--mpileup 1" : samtools mpileup -q 1 -f ~/hg19.fa normal.rmdup.bam tumor.rmdup.bam \| java -Xmx4g -jar ~/VarScan.v2.3.3.jar copynumber --mpileup 1 > output If this doesn't help, I believe the issue occurs at a position where you have 0x coverage... those usually aren't present in pileup files. Please send me the mpileup lines for chrM from positions 16180- 16190 and I'll take a look. Feel free to post here or e-mail to dkoboldt (at) genome [dot] wustl [dot] edu

12-23-2013, 10:01 PM	#3
lovenlong Member Location: Guangzhou, CN Join Date: Jan 2013 Posts: 16	Hi, all ! Though this thread had been posted about one year ago, I'd like to share my recent experience in meeting the same error in using Varscan2. The causal thing would be the inconsistent of samtools version in generating BAMs and producing mpileup files. In my case, input BAMs were generated by samtools-0.1.18 in weeks ago, while now the samtools version in my PATH is 0.1.19, and the error arised if I don't specify to use the samtools-0.1.18 in generating pileup file. Thanks a bunch!