CNV analysis for tumor panel

Question

[J.F.Jiang]

Hi all,

In order to evaluate the drug response of patient, we want to analyze the CNV status of target region.

The panel was developed and synthetized by IDT.

I am wondering if there is any tool to define the CNV status of the target region, such as MET/EGFR/HER2, beside GATK, FREEC, XHMM, since we find the accuracy problem while calling process.

Thanks

Junfeng

[Kevin Blighe]

There are many dozens of copy number calling programs. FREEC and XHMM are actually regarded as among the best. What, exactly, is your input data? - genome sequencing, exome / targeted sequencing, microarray, or PCR?

[J.F.Jiang]

the input data is target sequencing using IDT Capture probe.

[Kevin Blighe]

Okay, I have had success using FREEC for Exome-seq, which can be regarded as 'genome-wide' target capture. Did you configure FREEC properly? - there are many parameters. Here is an example config file that I used for an Exome-seq sample (Mus musculus):

[general]
BedGraphOutput=TRUE
bedtools=/Programs/bedtools2/bin/bedtools
breakPointThreshold=0.8
breakPointType=2
chrFiles=mm10/chromosomes/
chrLenFile=mm10/mm10.fasta.fai
SNPfile=mm10/mm10_dbSNP137.ucsc.freec.txt
coefficientOfVariation=0.06
contamination=0
contaminationAdjustment=TRUE
degree=3
forceGCcontentNormalization=1
#gemMappabilityFile=mm10/GRCm38_68_mm10.gem
#minMappabilityPerWindow=0.85
intercept=1
minCNAlength=3
minExpectedGC=0.35
maxExpectedGC=0.55
minimalSubclonePresence=0.3
maxThreads=3
numberOfProcesses=2
noisyData=TRUE
outputDir=Out/FREEC/
ploidy=2
printNA=TRUE
readCountThreshold=10
samtools=/Programs/samtools-1.9/samtools
sex=XY
step=10000
telocentromeric=50000
#uniqueMatch=TRUE
window=50000

[sample]
inputFormat=BAM
mateFile=Aligned_Sorted_PCRDuped_FiltMAPQ.bam
mateOrientation=FR

[J.F.Jiang]

Thanks,

I tried both FREEC and CNVKit. Both work well, but sometimes came with some bugs, especially depth calculation of FREEC.

Answer 1

[Eric T.]

CNVkit tends to work well on tumor panels and is easy to use.