Hi,
Im using VarScan.v2.3.2 to do CNV analysis on HighSeq exome data from tumor-normal pairs. When running copynumber, all appears ok, but when filtering the resulting .copynumber file using Copycaller, I get an error ("Parsing Exception", please see below).
When running VarScan copynumber like this:
java -jar /VarScan.v2.3.2.jar copynumber $NOR $TUM $BASENAME
I got the following output:
########
Normal Pileup: /177_1N.prmdup.realign.recal_sorted.mpileup
Tumor Pileup: /177_1T.prmdup.realign.recal_sorted.mpileup
Min coverage: 10
Min avg qual: 15
P-value thresh: 0.01
Not resetting normal file because chrM < chrY
561343988 positions in tumor
557785077 positions shared in normal
38214383 had sufficient coverage for comparison
482476 raw copynumber segments with size > 10
474997 good copynumber segments with depth > 10
##########
So we have an error stating "Not resetting normal file because chrM < chrY".
I saw an answer that dkobolt had given regarding this error message saying that "This is just a warning printed by VarScan as it's simultaneously parsing normal and tumor files. As long as your output files contain all of the chromosomes that you expect, you can safely ignore it."
So I double checked and all chromosomes are present in the copynumber file.
Then I ran Copycaller like this:
java -jar /VarScan.v2.3.2.jar copyCaller $IN --output-file ${IN}.called
The output I get from the CopyCaller is the following:
#####################
Min coverage: 20
Reading input from /177_1T.copynumber
Parsing Exception on line:
chr1 10010 10109 100 30,4 28,4 -0,097 51,0
For input string: "30,4"
Error parsing input: null
java.lang.NullPointerException
at net.sf.varscan.CopyCaller.<init>(CopyCaller.java:293)
at net.sf.varscan.VarScan.copyCaller(VarScan.java:344)
at net.sf.varscan.VarScan.main(VarScan.java:173)
##################
I am not sure what is wrong with this line, however when looking at the output from "copynumber" (please see below for a sample) I noticed that the GC content sometimes exceeds 100, please see attached picture.
##################
chrom chr_start chr_stop num_positions normal_depth tumor_depth log2_ratio gc_content
chr1 10010 10109 100 30,4 28,4 -0,097 51,0
chr1 10110 10209 100 23,4 21,4 -0,132 51,0
chr1 10210 10240 31 14,3 11,9 -0,260 51,6
chr1 10359 10458 100 20,7 14,1 -0,556 51,0
chr1 12202 12226 25 10,2 2,0 -2,350 48,0
chr1 13425 13438 14 10,0 5,9 -0,754 50,0
chr1 69005 69104 100 22,5 24,1 0,098 40,0
chr1 69105 69204 100 41,9 38,2 -0,133 77,0
chr1 69205 69304 100 74,6 70,9 -0,074 127,0
chr1 69305 69404 100 42,5 45,8 0,108 171,0
chr1 69405 69504 100 20,0 20,1 0,003 216,0
chr1 69505 69604 100 26,9 22,2 -0,277 265,0
chr1 69605 69704 100 66,4 64,2 -0,049 308,0
chr1 69705 69804 100 86,8 83,1 -0,064 42,0
chr1 69805 69904 100 73,5 71,1 -0,047 78,0
chr1 69905 70004 100 55,7 49,0 -0,185 119,0
chr1 70005 70043 39 22,4 23,9 0,096 25,6
chr1 367991 368039 49 11,2 5,3 -1,078 51,0
################
Any feedback would be greatly appreciated!
Thank you in advance.
Im using VarScan.v2.3.2 to do CNV analysis on HighSeq exome data from tumor-normal pairs. When running copynumber, all appears ok, but when filtering the resulting .copynumber file using Copycaller, I get an error ("Parsing Exception", please see below).
When running VarScan copynumber like this:
java -jar /VarScan.v2.3.2.jar copynumber $NOR $TUM $BASENAME
I got the following output:
########
Normal Pileup: /177_1N.prmdup.realign.recal_sorted.mpileup
Tumor Pileup: /177_1T.prmdup.realign.recal_sorted.mpileup
Min coverage: 10
Min avg qual: 15
P-value thresh: 0.01
Not resetting normal file because chrM < chrY
561343988 positions in tumor
557785077 positions shared in normal
38214383 had sufficient coverage for comparison
482476 raw copynumber segments with size > 10
474997 good copynumber segments with depth > 10
##########
So we have an error stating "Not resetting normal file because chrM < chrY".
I saw an answer that dkobolt had given regarding this error message saying that "This is just a warning printed by VarScan as it's simultaneously parsing normal and tumor files. As long as your output files contain all of the chromosomes that you expect, you can safely ignore it."
So I double checked and all chromosomes are present in the copynumber file.
Then I ran Copycaller like this:
java -jar /VarScan.v2.3.2.jar copyCaller $IN --output-file ${IN}.called
The output I get from the CopyCaller is the following:
#####################
Min coverage: 20
Reading input from /177_1T.copynumber
Parsing Exception on line:
chr1 10010 10109 100 30,4 28,4 -0,097 51,0
For input string: "30,4"
Error parsing input: null
java.lang.NullPointerException
at net.sf.varscan.CopyCaller.<init>(CopyCaller.java:293)
at net.sf.varscan.VarScan.copyCaller(VarScan.java:344)
at net.sf.varscan.VarScan.main(VarScan.java:173)
##################
I am not sure what is wrong with this line, however when looking at the output from "copynumber" (please see below for a sample) I noticed that the GC content sometimes exceeds 100, please see attached picture.
##################
chrom chr_start chr_stop num_positions normal_depth tumor_depth log2_ratio gc_content
chr1 10010 10109 100 30,4 28,4 -0,097 51,0
chr1 10110 10209 100 23,4 21,4 -0,132 51,0
chr1 10210 10240 31 14,3 11,9 -0,260 51,6
chr1 10359 10458 100 20,7 14,1 -0,556 51,0
chr1 12202 12226 25 10,2 2,0 -2,350 48,0
chr1 13425 13438 14 10,0 5,9 -0,754 50,0
chr1 69005 69104 100 22,5 24,1 0,098 40,0
chr1 69105 69204 100 41,9 38,2 -0,133 77,0
chr1 69205 69304 100 74,6 70,9 -0,074 127,0
chr1 69305 69404 100 42,5 45,8 0,108 171,0
chr1 69405 69504 100 20,0 20,1 0,003 216,0
chr1 69505 69604 100 26,9 22,2 -0,277 265,0
chr1 69605 69704 100 66,4 64,2 -0,049 308,0
chr1 69705 69804 100 86,8 83,1 -0,064 42,0
chr1 69805 69904 100 73,5 71,1 -0,047 78,0
chr1 69905 70004 100 55,7 49,0 -0,185 119,0
chr1 70005 70043 39 22,4 23,9 0,096 25,6
chr1 367991 368039 49 11,2 5,3 -1,078 51,0
################
Any feedback would be greatly appreciated!
Thank you in advance.
Comment