[BioC] Analyzing mulitple tissues

[David Kipling]

Hi Naomi,

I am glad you raised this issue of lack of replication.  I fully take 
on board your point.  Something is niggling me, however, so hopefully 
someone with a better stats background than me can help.

What puzzles me is that limma will allow you to run experiments with no 
replication *and* will return t-statistics for something like a 3x1 
comparison.   I don't generally run experiments without replication (he 
says quickly, defending himself!) but I've just tried a mock experiment 
that is made up of 3 states with 4, 3, and 1 degrees of replication 
(see snippet below).  Ordering by absolute(M) gives a ranking that is 
related to, but clearly distinct, from ranking by absolute(t statistic) 
or p-value.

Out of curiosity, what is limma doing here and how should one interpret 
these t stats/p-values (if indeed one should!)?  Are they any use over 
simple M values?

Regards

David



######################

#    Based on an example in the limmaUsersGuide
data1 <- ReadAffy()
eset <- rma(data1)

#   Note the single chip in group 2
design <- model.matrix(~ -1+factor(c(1,1,1,1,3,3,3,2)))
colnames(design) <- c("group1", "group2", "group3")

fit <- lmFit(eset, design)

#   Contrast 2 is a 1 v 3-chip comparison
contrast.matrix <- makeContrasts(group2-group1, group3-group2, 
group3-group1, levels=design)
fit2 <- contrasts.fit(fit, contrast.matrix)
fit2 <- eBayes(fit2)
topTable(fit2, coef=2, adjust="none")

########################


Prof David Kipling
Department of Pathology
School of Medicine
Cardiff University
Heath Park
Cardiff CF14 4XN

Tel:  029 2074 4847
Email:  KiplingD at cardiff.ac.uk
On 6 Jun 2005, at 15:22, Naomi Altman wrote:

> Biological inference implies that the "signal" can be observed above 
> the biological variation.  If you have no biological replicates, you 
> cannot determine if your signal is higher than the biological 
> variation.
>
> So, there is no statistically valid means of analyzing your data that 
> improves on an arbitrary choice of "fold difference", such as 2-fold 
> difference.
>
>
> Naomi S. Altman                                814-865-3791 (voice)
> Associate Professor
> Bioinformatics Consulting Center
> Dept. of Statistics                              814-863-7114 (fax)
> Penn State University                         814-865-1348 (Statistics)
> University Park, PA 16802-2111