[BioC] getting normexp (and log2 transformed) values - not further-processed M values

Gordon K Smyth smyth at wehi.EDU.AU
Wed Oct 7 00:25:42 CEST 2009

Dear Alana,

Since you want to do your own customized analysis of the data, which is 
different from what limmaGUI provides, you will find it easier to use the 
command-line limma package rather than limmaGUI.  All objects created will 
be quite transparent.

If you want to output values from an RGList object to a text file, you can 




However I find it hard to believe this is a good idea.  Better to find out 
how to do what you want in R.  I suspect you're making things harder for 
yourself than it needs to be.

Best wishes

> Date: Mon,  5 Oct 2009 19:36:38 -0400 (EDT)
> From: <adoty at umd.edu>
> Subject: [BioC] getting normexp (and log2 transformed) values - not
> 	further-processed M values
> To: bioconductor at stat.math.ethz.ch
> Message-ID: <20091005193638.ATB49372 at po5.mail.umd.edu>
> Content-Type: text/plain; charset="us-ascii"
> Dear Scott et al.
> Thank you very much for the information. After reading both the 
> limma.pdf and the usersguide.pdf (and playing around with my Windows R 
> console), I now know that by typing "fix(RG)" in limmaGUI's "Evaluate R 
> Code" tab, I can get the RG object to display itself in some form...
> I do, though, have 3 remaining (and hopefully final) questions on this topic:
> 1) what's the difference between RG and RGList?  I see them both used in 
> the help/example files and I can't tell which one has the output I'd 
> like from "backgroundCorrect" (-again, I'd like to obtain the normexp 
> backgrounded-corrected values, using an offset, for both the R and G 
> channels from each of my 5 arrays)
> 2) will the values in whichever of these "objects" I need [RG or RGList] 
> be log2 transformed already?  I'm not sure if this is automaticaly done 
> or not...
> 3) when I look at the corrected data with "fix(RG)", it's in matrix 
> form, right?  How can I get the data into, say, a form readable by 
> Excel?  I think I can get some useful output with "show(RG)," too - but 
> in this form it would require a lot of copying/pasting to get into Excel 
> (and by default, not all the lines are shown).
> Thank you VERY much for any help someone may be able to provide.  I am 
> trying to learn what I can on my own! - but it's going a bit more slowly 
> than I'd like some answers by...
> Sincerely,
> Alana Doty
> University of Maryland
> -------------- next part --------------
> An embedded message was scrubbed...
> From: Sean Davis <seandavi at gmail.com>
> Subject: Re: [BioC] getting normexp (and log2 transformed) values - not
> 	further-processed M values
> Date: Mon, 5 Oct 2009 07:07:14 -0400
> Size: 3187
> URL: <https://stat.ethz.ch/pipermail/bioconductor/attachments/20091005/26329477/attachment-0001.eml>
> ------------------------------

More information about the Bioconductor mailing list