phyloWGS compatibility

[etal]

phyloWGS compatibility

[etal]

I agree this would be a useful feature.

1. Should the exported CNV rows be all segments, or just the copy number variant (non-diploid) regions?
2. In column 4, do the maternal and paternal copy numbers need to be the integer value within each subclone (meaning CNVkit needs to figure out the subclone population structure), or a floating-point value where CNVkit doesn't need to have figured out the subclone structure yet?

[jungminchoilab]

This is an example cnv_file provided by the phyloWGS.

cnv a d ssms
c0 66023,50883,62757,36056,58777 126755,100469,121941,71263,115417 s2,1,2;s4,0,1

1. I believe this in just the copy number variants.
2. It seems like the copy numbers need to be the integer.

PhyloWGS seems to support the outputs of cgpBattenberg and TITAN and I was thinking it will be great if I can use the output from the CNVkit.

[etal]

It looks like phyloWGS input requires first using another program to infer subclone structure, which CNVkit doesn't do itself and wouldn't be trivial to add. So there will need to be another step, e.g. first running THetA2 or BubbleTree (see recent efforts in chapmanb/bcbio-nextgen), then converting the output of either of those programs to the phyloWGS input format.

Would it be helpful if CNVkit could generate the phyloWGS input file given one or more .cns files and a corresponding number of subclone fractions? This would still require you to run another program before phyloWGS, but it will decouple that choice of program a little, and allow you some more control over the selection of subclones and their fractions.

[etal]

Looping in @chapmanb -- this issue might be addressed more directly in bcbio-nextgen.

[wjaratlerdsiri]

Definitely useful for phylowgs compatibility!!! if cnvkit likes to be on top of CNV analyses.

James