[BioC] Re: Bioconductor Digest, Vol 12, Issue 28
Tarca Adi Laurentiu
ltarca at rsvs.ulaval.ca
Thu Feb 19 18:14:12 MET 2004
At 23:09 2004-02-18, you wrote:
>Date: Wed, 18 Feb 2004 16:25:36 -0600
>From: Joyce Gu <jwgu at bcm.tmc.edu>
>Subject: [BioC] Mising data
>To: bioconductor at stat.math.ethz.ch
>Message-ID: <4033E6B9 at webmail.bcm.tmc.edu>
>Content-Type: text/plain; charset="ISO-8859-1"
>I am using marrayInput packages to do my data analysis. After I read into my
>data with read.marrayRaw function. I found that lots of my data is labelled
>"NA". I use name.Gf="Ch1 Median",name.Rf="Ch2 Median",name.Gb=" Ch1 B
>Median",name.Rb="Ch2 B Median" command, I am wondering how marrayClass
>transformed data. Is still M vs A or what.
>I want to normalize my data with this packages, then export to do further
>analysis with other software.
>Any explanation is greatly appreciated about this algorithm
First thing to do is to verify that your files are ok and that the
variables names (e.g. " Ch1 B
Median") matches exactly those within your files.
You may start by reading in data from one of your files like for e.g.:
Then to read the file with the read.table function like:
data<- read.table(fname, header=TRUE, sep="\t", dec=".")
Now see the names of variables in your file like
You can check also the integrity your data looking at the values of "data"
>>I am wondering how marrayClass transformed data. Is still M vs A or what.
The object, (lets call it rawdata) of class marrayRaw that you obtain with
read.marray function will contain in the slots
maRf, maGf ...etc, exactly the same values you have in your files in the
columns Ch2 Median, Ch1 Median ..etc, so they are not
at all transformed. However, the A and M values (you may retrieve from the
slots maA, and maM) are additional, being computed
form maRf, maRb, maGf, maGb.
>>I want to normalize my data with this packages
For normalization you may use maNormMain function, after loading the
marrayNorm library. You may have details on it
with "? maNormMain".
A print-tip loess normalization may be simply done like
normdata<-maNormMain(rawdata, f.loc=list(maNormLoess() ))
>> then export to do further analysis with other software
You may retrieve any information from the nomalized object, normdata,
using the appropriate methods of the class
marrayNorm it belongs to. For eg maM(normdata) will give you a matrix with
the normalized M values
that you may further print into files with the "write.table" function for e.g.
More information about the Bioconductor