[BioC] FW: limma for 2 color data

[Sean Davis]

On Tue, Apr 15, 2008 at 7:20 AM, Afsaneh <s.maleki-dizaji at dcs.shef.ac.uk> wrote:
>
>
>
>
>  From: Afsaneh [mailto:s.maleki-dizaji at dcs.shef.ac.uk]
>  Sent: 15 April 2008 11:23
>  To: 'wettenhall at wehi.edu.au'
>  Subject: limma for 2 color data
>
>
>
>  Dear James,
>
>  I  was trying to use limmaPackge,
>
>
>
>  Here is my description file, at the moment I have got only one experiment.
>
>
>  SlideNumber
>
>  FileNameCy5
>
>  FileNameCy3
>
>  Cy3
>
>  Cy5
>
>
>  1
>
>  Imagene2.txt
>
>  Inagene1.txt
>
>  Wt
>
>  Mutant
>
>
>
>
>
>  The problem is after normalization when I try to run
>
>  fit = lmFit(MA,design= c(1))     ok
>
>
>
>  > topTable(fit)
>
>  Error in dim(data) <- dim : attempt to set an attribute on NULL
>
>
>
>  efit <- eBayes(fit)  ??
>
>
>
>  Error in ebayes(fit = fit, proportion = proportion, stdev.coef.lim =
>  stdev.coef.lim) :
>
>   No residual degrees of freedom in linear model fits

Hi, Afsaneh.

You will not be able to do statistics with only one sample.  The best
you can do is to order your genes by fold change.  The error message
above essentially tells you that, but in stats terms.

Sean



>   tops = topTable(efit,coef=1,adjust='fdr',sort.by='B',number = 50000)
>
>  I would really appreciate if let me know what is the problem.