Apply Idr On List Of Read Counts

Question

[14134125465346445]

Does anybody know how to apply the IDR code on lists of read counts per peak, instead of the usual narrowPeak format:

peak-1    20
peak-2    25
peak-3    23
...

So that using IDR and my peak counts, I get a threshold of the minimum peak intensity for my samples?

I suppose the best version to use is the one explained below, as seen on this url:

http://sites.google.com/site/anshulkundaje/projects/idr

(1) batch-consistency-analysis.r

----------------
GENERAL USAGE:
----------------
Rscript batch-consistency-analysis.r [peakfile1] [peakfile2] [peak.half.width] [outfile.prefix] [min.overlap.ratio] [is.broadpeak] [ranking.measure]

Typical usage for SPP peak caller peaks
Rscript batch-consistency-analysis.r [peakfile1] [peakfile2] -1 [outfile.prefix] 0 F q.value

Typical usage for MACSv2 peak caller peaks
Rscript batch-consistency-analysis.r [peakfile1] [peakfile2] -1 [outfile.prefix] 0 F p.value

[peakfile1] and [peakfile2] are the peak calls for the pair of replicates in narrowPeak format. They must be uncompressed files.
e.g. /peaks/reps/chipSampleRep1_VS_controlSampleRep0.narrowPeak AND /peaks/reps/chipSampleRep2_VS_controlSampleRep0.narrowPeak

Any ideas?

[Gjain]

may be this can help Irreproducible Discovery Rate and http://cran.r-project.org/web/packages/idr/idr.pdf

Answer 1

[Ying W]

I think you need a bit more information to call IDR. It is possible to convert your file of peaks to to a narrowPeak-like format by adding artificial non-overlapping chromosome locations however I think IDR works best with SPP output. You could try posting to their google groups page, the author (Anshul) is pretty active on there.