Diffbind: Peak score normalization
2
0
Entering edit mode
medinaale • 0
@medinaale-7249
Last seen 9.2 years ago
Canada

Dear Rory

 

First I wan to thank you for making Diffbind an easy access tool. I have been using it a lot and I have a couple of questions.

 

I have been looking for a description on how the peak score is taken into account by default when loading the data using dba.

 

The peaks are in narrow format and I load the data with this command

EC.data <- dba(sampleSheet=EC.table, minOverlap=1)

 

Then I get the merged peaks using

 

peaks.init.ranges<-dba.peakset(EC.data, bRetrieve=TRUE)

 

But the values I can see are not corresponding to the peak -log(10)pvalue that was in the narrowPeak file, it seams to have been normalized, can you please explain me how this normalization is done?

I thought the -log(10)value from the narrowPeaks file was directly begin used by default.

 

 

Thanks

 

Alejandra Medina

 

 

Genetics & Genome Biology Program,

SickKids Research Institute

Peter Gilgan Centre for Research and Learning

686 Bay Street  | Room 14.9713

Toronto, ON, Canada. M5G 0A4

Courier address (72 Elm St. M5G 0A4)

Lab Phone: 416.813.7654 | Ext. 328699

Website: wilsonlab.org

normalization chipseq diffbind • 2.0k views
ADD COMMENT
0
Entering edit mode
Rory Stark ★ 5.1k
@rory-stark-5741
Last seen 8 weeks ago
Cambridge, UK

Hi Alejandra-

The scores are normalized to a [0.1] scale. Each time a peak set is added, the scores are divided by the maximum score.This way scores from different peak caller using different scales can be compared. 

If you look at the documentation for the scoreCol parameter of dba.peakset(), it says that specifying a value of 0 will disable the normalization. I haven't tested that recently, and I'm not sure if it will also work in call to dba() (when loading a sample sheet). If you are interested in running without normalized peak scores let me know and I can verify that bit works properly...

Cheers-

Rory

ADD COMMENT
0
Entering edit mode
medinaale • 0
@medinaale-7249
Last seen 9.2 years ago
Canada

Thanks, 

I just wanted to know how the normalization was being done, I actually think is better to normalize it since my datasets are coming from various sources. 

 

Cheers

Alejandra

ADD COMMENT

Login before adding your answer.

Traffic: 699 users visited in the last hour
Help About
FAQ
Access RSS
API
Stats

Use of this site constitutes acceptance of our User Agreement and Privacy Policy.

Powered by the version 2.3.6