Dear list,
I have encountered following problem and do not know how to solve it:
After processing my data with RMA and filtering genes, that are not of
interest I have performed SAM analysis. Here are the steps I took:
cell <- ReadAffy()
RMAcell <- rma(cell)
after filtering my data set, I have the following:
filtered
Expression Set (exprSet) with
2271 genes
20 samples
phenoData object with 1 variables and 20 cases
varLabels
sample: arbitrary numbering
I now perform SAM anaylsis like this to extract the genes of interest
(I
want to have about 120 called genes):
cl <- c(0,0,0,0,0,0,0,0,0,0,1,1,1,1,1,1,1,1,1,1)
where 0 stands for the probes of group 0, 1 for the samples of group
1.
c <- seq(1.3, 1.36, 0.01)
sam(filtered, cl, delta=c, rand =123)
SAM Analysis for the Two-Class Unpaired Case Assuming Unequal
Variances
Delta p0 False Called FDR
1 1.30 0.464 7.44 127 0.0272
2 1.31 0.464 7.30 126 0.0269
3 1.32 0.464 6.71 120 0.0260
4 1.33 0.464 6.41 117 0.0254
5 1.34 0.464 5.50 109 0.0234
6 1.35 0.464 5.50 109 0.0234
7 1.36 0.464 5.07 103 0.0229
now I summarize my results for delta= 1.32 :
sumRMA <- summary(sam(filtered,cl, delta=1.32, rand=123))
and want to extract those 120 genes like this:
top120RMA<-filtered[sumRMA at row.sig.genes,]
top120RMA
Expression Set (exprSet) with
0 genes
20 samples
phenoData object with 1 variables and 20 cases
varLabels
sample: arbitrary numbering
but as you can see, my expression set contains 0 genes out of my 20
samples.
If I try the same steps but with GCRMA instead of RMA, everything
works out
well (of course with a different value for delta).
Every kind of help would be very much appreciated.
Thanks in advance!
With best regards,
gregor
