[BioC] DESeq on CCAT identified chipseq peaks

[QAMRA Aditi (GIS)]

Thank you so much for explaining all of it so well.
I read an answer you just gave about using the summits option in dba.counts - What version of diffbind is that ? I am using version 1.8.5 in Rversion 3.0.2 and I can't use the summits option.

Aditi

-----Original Message-----
From: Rory Stark [mailto:Rory.Stark at cruk.cam.ac.uk]
Sent: Friday, May 16, 2014 10:17 PM
To: QAMRA Aditi (GIS)
Cc: bioconductor at r-project.org
Subject: Re: [BioC] DESeq on CCAT identified chipseq peaks

Hello Aditi-

What you want is to use a "matched" design. There is a good explanation of this design (in a differential expression context) in the edgeR vignette.
Basically, the matched tumour-normal pairs are going to have certain similarities to each other as they each come from the same patient. A matched design will model this to detect consistent differences in enrichment between tumor and normal that is independent of individual patients.

You can analyze a matched design by setting up the contrast with
block=DBA_REPLICATE:

> h3k4me3_counts = dba.contrast(h3k4me3_counts,
>categories=DBA_CONDITION,
>block=DBA_REPLICATE)
> h3k4me3_counts = dba.analyze(h3k4me3_counts, method=DBA_DESEQ2)

You'll see that two analyses are run (unmatched and matched):

> h3k4me3_counts

Is is useful to look at the MA plot:

> dba.plotMA(h3k4me3_counts, method=DBA_DESEQ2_BLOCK)

You can get the list of all the sites with statistics relating to how confidently they can be identified as being differentially enriched:

> matchedReport = dba.report(h3k4me3_counts, method=DBA_DESEQ2_BLOCK,
> th=1)

Cheers-
Rory


On 16/05/2014 06:54, "QAMRA Aditi (GIS)" <qamraa99 at gis.a-star.edu.sg>
wrote:

>Hi Dr. Rory,
>
>I understand now. Thank you !
>
>A last question (hopefully) - Can you explain a little more on how the
>use of a blocking factor works in the case of matched normal tumor
>pairs ? Does it mean that using the DBA_REPLICATE condition as a
>blocking factor in such a case adjusts (?) and removes any sort of
>batch effects between replicates ?
>
>Thanks !
>Aditi
>________________________________________


-------------------------------
This e-mail and any attachments are only for the use of the intended recipient and may be confidential and/or privileged. If you are not the recipient, please delete it or notify the sender immediately. Please do not copy or use it for any purpose or disclose the contents to any other person as it may be an offence under the Official Secrets Act.