[BioC] PCA methods for probe level data

[James W. MacDonald]

I don't know of any packages designed for probe-level analysis, nor do I 
know what you mean by SFP detection.

However, if you just want to use the probe values for analysis you can 
simply  background correct and normalize the data and then analyze. For 
some packages you might want to convert your AffyBatch to an 
ExpressionSet (see Martin Morgan's email from yesterday for an example).

As for plotPCA(), this is just a thin wrapper around prcomp() from base 
R. Something like
pca <- prcomp(t(exprs(abatch)))
plot(pca$x[,1:2])

will do essentially the same. You might need to adjust the 'pch' and 
'col' arguments to plot() in order to tell which sample is which, but 
otherwise that's all plotPCA() is really doing.

Best,

Jim

zhi.zhang at syngenta.com wrote:
> Hello All,
>
> I am using Affy GeneChip for SFP detection and only interested in the
> probe-level data. Who knows there are some PCA packages or tools that
> can analyze probe-level data. I tried PCAplot in affycoretools package,
> but seems it only takes expression data. Any suggestion? Thanks in
> advance.
>
> Best,
>
>
> Zhi Zhang
> Lead Selection
> Applied Genomics
>
> Syngenta Biotechnology, Inc.
> 3054 Cornwallis Road
> Research Triangle Park, NC 27709-2257
>
> Tel: (919)765-5052
> Fax: (919)541-8585
>
>
>
> 	[[alternative HTML version deleted]]
>
> _______________________________________________
> Bioconductor mailing list
> Bioconductor at stat.math.ethz.ch
> https://stat.ethz.ch/mailman/listinfo/bioconductor
> Search the archives: http://news.gmane.org/gmane.science.biology.informatics.conductor
>