[BioC] Gene filtering for differential expression (limma)

michael watson (IAH-C) michael.watson at bbsrc.ac.uk
Thu Jul 21 11:56:09 CEST 2005

Gordon's right, if you post an example of your targets file we will have
a better idea of what you mean

>>Initial quality control and
>>spot filtering are performed in image analysis program.

Personally, I wouldn't recommend doing this.  The way R works, it's
better to have all your data points present in all files.  I leave all
data in and flag up bad spots, and remove them at the end of the
analysis, not the beginning


-----Original Message-----
From: bioconductor-bounces at stat.math.ethz.ch
[mailto:bioconductor-bounces at stat.math.ethz.ch] On Behalf Of Gordon
Sent: 21 July 2005 10:30
To: Spela Baebler
Cc: bioconductor at stat.math.ethz.ch
Subject: Re: [BioC] Gene filtering for differential expression (limma)

At 06:10 PM 21/07/2005, =?iso-8859-2?Q?=A9pela_Baebler?= wrote:
>Nice to get a reply:-)
>We now have a problem which is more statistics than R. We have a 
>with creating a design and contrast matrix.
>If we just want to compare vithin one variety, it is easy design <- 
>c(1,-1,1,-1,1) where we just take into the account the dye swaps.
>We couldn't find any example that would fit our experimental design in
>Limma users Guide, since we don't have a common reference for all
>So do you have a suggestion for contrast matrix if we want to compare
>4 varieties?

Perhaps you need to read the documentation more carefully because there
examples in the User's Guide of two colour experiments without a common 
reference. For example, Section 9.4 "Direct Designs" considers an
with 3 treatments.

Do you mean "design matrix" rather than "contrast matrix", or are you 
implying you already have a design matrix?

Have you tried reading in your targets file and using modelMatrix()?

>Waiting for answers...

To make good use of this mailing list you need to read the documentation

carefully, get as far as you can on your own, and then ask questions
are specific as possible. Just waiting for complete answers is not on.

BTW, what does your question have to do with "gene filtering"?


>P.S:: Is there someone using TIGR potato microarrays out there?
>-----Original Message-----
>From: Gordon K Smyth [mailto:smyth at wehi.EDU.AU]
>Sent: Thursday, July 21, 2005 12:58 AM
>To: (c)pela Baebler
>Cc: bioconductor at stat.math.ethz.ch
>Subject: [BioC] Gene filtering for differential expression (limma)
> > Date: Tue, 19 Jul 2005 10:53:37 +0200
> > From: ?pela Baebler <Spela.Baebler at nib.si>
> > Subject: [BioC] Gene filtering for differential expression (limma)
> > To: <bioconductor at stat.math.ethz.ch>
> >
> > Hello everyone!
> >
> > We are struggling to implement R for the analysis of two color cDNA
> microarrays.
> >
> > Each microarray is hybridised with treated control plant RNA. There 
> > are
> 5 biological repetitions,
> > they also include dye swaps. We would like to compare 4 groups of
> plants (cultivars) having the
> > same treatment, meaning altogether the experiment has 20
> microarrays.  Initial quality control and
> > spot filtering are performed in image analysis program.
> >
> > We are trying to do further analysis with limma package, although we
> are not experienced R users.
> >
> > We are wondering is limma the appropriate package for such an
> experimental outline, or we should
> > try any other packages? Which ones would you recommend?
>limma is certainly intended to address this sort of experiment.  Have 
>run into problems or are
>you just trying to get an idea of the alternatives?
> > Thanks for any answers/comments,
> >
> > Spela Baebler
> >
> > National Institute of Biology
> > Dept. of Plant Physiology and Biotechnology
> > Ljubljana

Bioconductor mailing list
Bioconductor at stat.math.ethz.ch

More information about the Bioconductor mailing list