Question: Diffbind results - non uniform width
gravatar for YaGalbi
17 months ago by
MRC Harwell Institute, Oxford, UK
YaGalbi20 wrote:

Hi all,

I am running diffbind using narrowPeaks output from macs2

When using dba.count, I am taking each summit and counting reads +/-250 from the summit so I'm expecting to see in the diffbind report that the width column only contains "501"

bindingmatrix <- dba.count(dbaobject, summits=250) # summit +/- 250 bases

However, around 2% of the rows in the report contain a range of widths up to ~1400

Can anyone explain why this is?

Thank you.

ADD COMMENTlink modified 17 months ago by Rory Stark2.6k • written 17 months ago by YaGalbi20
gravatar for Rory Stark
17 months ago by
Rory Stark2.6k
CRUK, Cambridge, UK
Rory Stark2.6k wrote:

If any of the 501bp intervals overlap (by at least one base), they will be merged into a wider peak. The 1400bp intervals would have been created because there are three peaks quite close together.

If you want to differentiate between very narrow peaks quite close together, you can set the summits parameter lower (100 or even 50).



ADD COMMENTlink written 17 months ago by Rory Stark2.6k

Thank you once again Rory :)

ADD REPLYlink written 17 months ago by YaGalbi20
Please log in to add an answer.


Use of this site constitutes acceptance of our User Agreement and Privacy Policy.
Powered by Biostar version 2.2.0
Traffic: 154 users visited in the last hour