# [BioC] Read single channel GenePix in limma [was: Analyze miRNA experiment in Bioconductor]

Daniel Brewer daniel.brewer at icr.ac.uk
Mon May 19 14:35:49 CEST 2008

```Not the most experienced in this area, but I think your design is pretty
much like section 8.5 in the Limma's user guide.

"There are two major ways in which this comparison can be made.
We can
1. create a design matrix which includes a coefficient for the mutant
vs wild type difference,
or
2. create a design matrix which includes separate coefficients for wild
type and mutant mice and then extract the difference as a contrast."

So in your situation you would have either
> design
plus   plusvsminus
a  1      0
b  1      0
c  1      0
d  1      0
e  1      1
f  1      1
g  1      1
> fit <- lmFit(eset, design)
> fit <- eBayes(fit)

or
plus   minus
a  1      0
b  1      0
c  1      0
d  1      0
e  0      1
f  0      1
g  0      1
> fit <- lmFit(eset, design)
> cont.matrix <- makeContrasts(plusvsminus=plus-minus, levels=design)
> fit2 <- contrasts.fit(fit, cont.matrix)
> fit2 <- eBayes(fit2)

Hope that helps.

Paul Geeleher wrote:
> Ok I'm close to having this all sorted and have the
> duplicateCorrelation() function working. For posterity I'd be happy to
> post a the code and detailed explanation from everything I've done to
> the mailing list. It should be useful for anyone doing similar MiRNA
> analysis in future. My last question is regarding the design matrix,
> I'm not sure if I'm creating it properly.
>
> I have 7 arrays, a, b, c, d, e, f and g. I want to measure
> differential expression of arrays a, b, c & d against e, f & g.
>
> In some of the tutorials in the limmaUsersGuide, this design matrix is
> simply created as follows:
>
> design <- c(-1, -1, -1, -1, 1, 1, 1)
>
> I've also seen it created using methods like this:
>
> pData <- data.frame(population = c('HER2+', 'HER2+', 'HER2+', 'HER2+',
> 'HER2-', 'HER2-', 'HER2-'))
> rownames(pData) <-  RG\$targets\$FileName
> design <- model.matrix(~factor(pData\$population))
>
>
> These two different methods give me very different p-values come the
> end of analysis and I'm wondering what exactly I should be doing?
>
>
> -Paul.
>
> On Thu, May 15, 2008 at 1:40 AM, Gordon K Smyth <smyth at wehi.edu.au> wrote:
>> Dear Paul,
>>
>> I have no experience with miRNA arrays, so cannot give you any specific
>>
>> With ordinary expression arrays, my practice is to keep the probes separate,
>> especially if they actually have differing sequences.  There's no good
>> summarisation method which feeds into the topTable format.  I convert to
>> genes or transcript only at the interpretation stage.  The only exception
>> are within array replicates which satisfy the (restrictive) assumptions of
>> the duplicateCorrelation() function.
>>
>> Best wishes
>> Gordon
>>
>> On Wed, 14 May 2008, Paul Geeleher wrote:
>>
>>> Hi Gordon,
>>>
>>> Thanks for you email. I've followed your steps and am getting some output
>>> now.
>>>
>>> One problem though. When should the summarization step occur? What I
>>> mean is that, between miRNA and control signals, my GPR file contains
>>> about 3000 entries and when I am done with analysis topTable will
>>> return all of these individually. But many of the miRNAs have multiple
>>> entries in the ".gpr" file. So how, and when, should I go about
>>> combining these into one value?
>>>
>>> -Paul
>
> _______________________________________________
> Bioconductor mailing list
> Bioconductor at stat.math.ethz.ch
> https://stat.ethz.ch/mailman/listinfo/bioconductor
> Search the archives: http://news.gmane.org/gmane.science.biology.informatics.conductor

--
**************************************************************

Daniel Brewer

Institute of Cancer Research
Molecular Carcinogenesis
MUCRC
Sutton, Surrey SM2 5NG
United Kingdom

Tel: +44 (0) 20 8722 4109
Fax: +44 (0) 20 8722 4141

Email: daniel.brewer at icr.ac.uk

**************************************************************

The Institute of Cancer Research: Royal Cancer Hospital, a charitable Company Limited by Guarantee, Registered in England under Company No. 534147 with its Registered Office at 123 Old Brompton Road, London SW7 3RP.

This e-mail message is confidential and for use by the a...{{dropped:2}}

```