[BioC] about the use of maW in marrayInput

Margherita atreeneedsaforest at tiscali.it
Wed Feb 11 16:47:50 MET 2004


Hello,

I am using marrayInput package to import microarray data from different 
platform sources and I have a couple of doubts about if data are 
modified and how as the marrayRaw object is created (I'll send separate 
emails for different questions):
- in what form should the maW value be? Is it a numeric value ranging 
0:1 or what? In the help I couldn't find any hint about how to use it.
In my datafiles there are columns with strings like "DETECTED" or 
"PASSED" or "LOW_SIGNAL" etc. and I would like to use these informations 
to flag good or bad spots, such that they are eliminated from subsequent 
normalization.
I have thought to catch these columns with read.marrayInfo and assign 
maW  "good" or "bad" on the basis of it. Also, in some cases I have 
different indicators for the two channels, so I have to check both to 
see if the spot is reliable.
I am trying something like:

#files being the list of datafiles
for (i in files)   assign(paste("info", i, sep=""),  read.marrayInfo(i, 
info.id=1:22, labels=1))

#and, for each of them, es slide A:
Agreen<-maInfo(infoA)[,21]
Ared<-maInfo(infoA)[,22]
AG<-as.vector(Agreen)
AR<-as.vector(Ared)
#assuming maW=1 for "good" spots and maW=0 for the "bad" ones
Aw<-rep(1, maNspots(data))
Aw[AG=="somewhat-bad"]<-0
Aw[AR=="some-else-bad"]<-0
slot(data, "maW")[,1]<-Aw

Is there a better way to do it?

Thank you to any that would help me

Margherita



More information about the Bioconductor mailing list