Varscan Copy Number Alterration Caller, Number Of Markers

Hi,

I'm currently working happily with the VarScan2 Copy Number Alteration caller for paired exome seq samples. But i'm still a bit confused about the "markers" in the number of markers output. Anyone knows if those are actual exomes or sequenced regions used as markers for the CBS or anything else i didn't get completely?

Best Ron

varscan

• 4,796 views • link

written 12.7 years ago by ron_veg •

I don't know if the markers are supposed to be placed on your gene of copy-number interest, or if you mark a relatively normal area as a sort of normalization.

I have been working happily with Varscan2 and have generated the copynumber data for 10 whole genomes, although I have yet to process it fully. I think you're referring to the Recurrent CNA Identification with CMDS? There is a readme at http://sourceforge.net/projects/cmds/files/ that has some information, but the only reference I see to markers is at the bottom of http://varscan.sourceforge.net/copy-number-calling.html

• 0 views • link

written 12.7 years ago by DoubleD

2 answers

I've used the somatic copy number caller in VarScan before, but I don't see a "markers" column.

There are the columns I see in the somatic output column:

 chrom    chr_start    chr_stop    num_positions    normal_depth    tumor_depth    log2_ratio

A description of these variables can be found on the VarScan website:

http://varscan.sourceforge.net/using-varscan.html#v2.3_copynumber

If "markers"="positions", then that variable refers to the length of the copy number alternation

• 0 views • link

written 12.7 years ago by Charles Warden

Sorry i should have specified this early.

I'm using CBS segemention from DNAcopy(Rpackage) to segment the raw output of VarScan2 CopyNumber. The mentioned Makers are from the CBS output and then post processed with the additional VarScan Merge Script.

• 0 views • link

written 12.7 years ago by ron_veg •

I could see this being necessary for the VarScan single-sample copy number calls (which I don't think are very useful), but the somatic copy number calls (meant for paired samples like you have) already defines regions with start and stop coordinates. This is actually one of the nice features of VarScan. I would recommend taking a look at those results (the link above is specifically for the somatic copy number caller).

• 0 views • link

written 12.7 years ago by Charles Warden