Help With The Dnacopy Number Package In R

Question

[ivivek_ngs]

Dear All,

I am using the Varscan 2 for my experiments to understand the variants and mutational profile in my tumor an IPS(induced pluripotent stem cells). I am doing the CNA calling using the pakcage DNACopy and using CBS for segmentation. I am trying to understand for each of my test sample(tumor or IPS) at individual chr level how does the profile looks like after plotting, if I can see any amplification or deletion at any chr and if the same thing corresponds to its IPS which is derived from a single clone from the tumor. To this when I use the R code , I see for my tumor and IPS the plot scales are different on the y- axis. I would like to have same scale for both tumor and IPS data. Can anyone suggest me how to do that? I am more interested in controlling the scale Y axis (log2ratio values of mean). I want a scale between -2 to 2 , but I see the scale between -4 to 4 in tumor and for two IPS it is -2 to 2 and-6 to 6. I want to have same scale for all the 3. first I want to have the scale between -2 to 2 and then play with the scale more to see the distinct pattern of the amplifications and deletions at each chr level for the genomic segments. I am attaching the piece of code I used. Any suggestions on changing the scale is welcome.

cn <- read.table("Desktop//CNA/N1vsLG_CNV_filtered.txt",header=T)
CNA.object <-CNA( genomdat = cn[,7], chrom = cn[,1], maploc = cn[,2], data.type = 'logratio',sampleid = "N1vsLG_CNV") ## cn[,5]`
CNA.smoothed <- smooth.CNA(CNA.object)
segs <- segment(CNA.smoothed, verbose=0, min.width=2)
segs2 = segs$output
pdf(file = "~/Desktop/tmp.pdf")
plot(segs, plot.type = "c")
dev.off()

Which should be the parameter I should use to obtain the similar scale for all different comparisons. Please advice.

Thanks

[Sean Davis]

Try plot(...., ylim=c(-2,2)).

[ivivek_ngs]

Yes thanks now it is working, I was taking the scale wrong

[Irsan]

If each of your tumors can be matched with an IPS sample, isnt it better to calculate the log2(tumor/IPS) and use that as input for DNAcopy ?

[ivivek_ngs]

They are actually log2 values only as far as I know. The VarScan uses the log2 values only for CNV analysis and then uses CBS with the DNAcopy tool.

Answer 1

[Sean Davis]

Just echoing my comment since it seemed to do the trick.

plot(...., ylim=c(-2,2))