SEQanswers

Go Back   SEQanswers > Applications Forums > RNA Sequencing



Similar Threads
Thread Thread Starter Forum Replies Last Post
MiSeq gDNA reads still fail "Kmer content" and "per base seq content" after trimming" ysnapus Illumina/Solexa 4 11-12-2014 07:25 AM
DEXSeq error in estimateDispersions: match.arg(start.method, c("log(y)", "mean")) fpadilla Bioinformatics 14 07-03-2013 02:11 PM
Relatively large proportion of "LOWDATA", "FAIL" of FPKM_status running cufflink ruben6um Bioinformatics 3 10-12-2011 12:39 AM
The position file formats ".clocs" and "_pos.txt"? Ist there any difference? elgor Illumina/Solexa 0 06-27-2011 07:55 AM
"Systems biology and administration" & "Genome generation: no engineering allowed" seb567 Bioinformatics 0 05-25-2010 12:19 PM

Reply
 
Thread Tools
Old 07-23-2015, 01:04 AM   #1
David [R]
Member
 
Location: France

Join Date: Jul 2012
Posts: 8
Default edgeR, "DGELIst" function, "group" parameter

Hi,

I am using edgeR for some RNASeq analysis. My question concerns the DGEList function and its "group" parameter. I realized that, whether doing group=mydesignfactor or group=rep(1,ncol(counts)), the cpm numbers that I get after
counts.TMM <- calcNormFactors(counts.DGEList)
cpm.TMM=cpm(counts.TMM, normalized.lib.sizes=TRUE)
are strictly identical in either case. The only difference between both ways is that the counts.TMM$samples$group column duly reflects what I put in in "group".

Is this normal? What is it that I am missing? Is the "group" nevertheless taken into account on the subsequent estimateGLMCommonDisp, estimateGLMTrendDisp, estimateGLMTagDisp, glmFit, glmLRT etc.?

Thanks in advance for your help.

Best,

David Rengel
David [R] is offline   Reply With Quote
Old 07-23-2015, 03:52 AM   #2
dpryan
Devon Ryan
 
Location: Freiburg, Germany

Join Date: Jul 2011
Posts: 3,479
Default

There's no reason that cpm should be dependent upon your experimental design (in fact, it shouldn't be). The various GLM function pay attention to the design, since that's where it's important.
dpryan is offline   Reply With Quote
Old 07-23-2015, 04:11 AM   #3
David [R]
Member
 
Location: France

Join Date: Jul 2012
Posts: 8
Default

Hi dpryan,

THanks for your reply. Yes, that is right. My only only doubt came from de fact that if normalized.lib.sizes=TRUE, then the cpm are calculated using the normalization factors, which I thought were somehow dependent on your experimental design, but it seems I was mistaken.

David
David [R] is offline   Reply With Quote
Old 07-23-2015, 04:18 AM   #4
dpryan
Devon Ryan
 
Location: Freiburg, Germany

Join Date: Jul 2011
Posts: 3,479
Default

Correct, normalization factors are independent from experimental design.
dpryan is offline   Reply With Quote
Old 07-23-2015, 04:29 AM   #5
David [R]
Member
 
Location: France

Join Date: Jul 2012
Posts: 8
Default

Thanks a lot!
David [R] is offline   Reply With Quote
Reply

Tags
edger, normalization, rnaseq

Thread Tools

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off




All times are GMT -8. The time now is 10:03 AM.


Powered by vBulletin® Version 3.8.9
Copyright ©2000 - 2018, vBulletin Solutions, Inc.
Single Sign On provided by vBSSO