Question

DiffBind concentration calculation

0

Entering edit mode

igor ▴ 40

@igor

Last seen 10 months ago

United States

When you retrieve the differentially bound sites with dba.report(), there is a "Conc" column that shows the mean read concentration over all the samples (log2 normalized ChIP read counts with control read counts subtracted). Is there a way to calculate those numbers for arbitrary regions (not just differentially bound)?

diffbind chipseq • 1.5k views

ADD COMMENT • link updated 8.8 years ago by Rory Stark ★ 5.2k • written 8.8 years ago by igor ▴ 40

score 0 · Answer 1 · 2015-07-09

0

Entering edit mode

Rory Stark ★ 5.2k

@rory-stark-5741

Last seen 13 days ago

Cambridge, UK

Hi Igor-

If you set th=1 in the call to dba.report(), you'll get the stats for all the regions.

Cheers-

Rory

ADD COMMENT • link 8.8 years ago Rory Stark ★ 5.2k

0

Entering edit mode

Yes, I know, but those are the regions as determined by DiffBind. My question is if I want to look at arbitrary regions. For example, regions that were found in a different experiment.

ADD REPLY • link 8.8 years ago igor ▴ 40

0

Entering edit mode

Hi,

You can't probe arbitrary regions on the fly. If you have a fixed list of regions, you can provide that when you count reads via the "peaks" option, as in

my_obj <- dba.count(...,peaks=myPeaks,...)

where peaks is a GRanges or other form of peak list (see the docs for details), then it will count and calculate everything on exactly those peaks.

Cheers

- Gord

ADD REPLY • link 8.8 years ago Gord Brown ▴ 650