[BioC] how to remove unwanted spots from topTable output?

[Marcelo Luiz de Laia]

Dear Dennis,

I has the some problem yasterday and I fix it with a sub set.

I group this blank in the end of the data file.

For example, from 1 to 2576 genes I have the "true" data. From 2577 to 2898 I have the blank genes.

When I execute the limma commands, I sub set it. 

MA<-normalizeWithinArrays(RG, layout)
MA<-normalizeBetweenArrays(MA,method="scale")
cor<-dupcor.series(MA$M[1:2576,],design,ndups=2,spacing=4)
fit<-lmFit(MA[1:2576,],design,ndups=2,spacing=4,correlation=cor$cor)
eb<-eBayes(fit)
TopTableresult<-topTable(eb,number=100,adjust="fdr")
write.table(TopTableresult,"result.txt",sep="\t")

I dont known if this is the best way for this, but, for me, it solve my problem, I think. My excuses if it is wrong.

Hope this help

Marcelo


Em Fri, 07 May 2004 10:46:49 -0700
Dennis Hazelett <hazelett at uoneuro.uoregon.edu> escreveu:

DH> I have a class of spots on my arrays called "blank" that I would like to 
DH> remove from topTable output. I didn't see an easy way to do this in any 
DH> of the options.
DH> 
DH> --Dennis