[BioC] Affy/annaffy question

[James W. MacDonald]

Hi Achilleas,

On 10/22/2010 5:11 PM, Achilleas Pitsillides wrote:
> Hi all,
>
> I was looking over an analysis that was done by someone in my lab in which
> the person used the aafTableInt on the raw data (AffyBatch object created by
> ReadAffy to save the expression data.
> So I was wondering what algorithm does aaTableInt use for summarizing the
> probes to probesets, so I looked into the code of
> aafTableInt(AffyBatch,probeids=featureNames(AffyBatch)) which really
> confused me because in effect it returned
> exprs(AffyBatch)[match(probeids,featureNames(AffyBatch)),]
> It was my understanding that exprs(AffyBatch) object returned the probe
> expression levels and not the probeset summarized expression data, am I
> missing something?

No, but it appears your colleague might be. There is nothing in annaffy 
to summarize the raw data - that task is handled primarily by the affy 
package.

I am a bit surprised that annaffy would accept an AffyBatch as input, 
but since the AffyBatch has the same parent class (eSet) as the 
ExpressionSet, maybe that is to be expected.

Anyway, I can't think of any reason to be feeding an AffyBatch into 
annaffy. The data need to be normalized, summarized, and comparisons 
made before you do anything involving annaffy.

>
>   I also noticed that the sum of the pm probes and mm probes is less than the
> nrows of the exprs(AffyBatch). Are the rest of the probes just controls?

No, the controls are mapped in the AffyBatch as well. What you are 
missing are the oligo(dT) probes that are primarily found on the very 
outside of the chip. These form the checkerboard pattern you can see if 
you zoom in on a CEL or DAT file using the affy software (or using 
image() within R, and selecting a subset of the data). The oligo(dT) 
probes are used by the scanner to align the image to the mask of the 
probe positions as one of the first steps in image processing.

Best,

Jim


>
> Thanks in advance,
> Achilleas
>
> 	[[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

-- 
James W. MacDonald, M.S.
Biostatistician
Douglas Lab
University of Michigan
Department of Human Genetics
5912 Buhl
1241 E. Catherine St.
Ann Arbor MI 48109-5618
734-615-7826
**********************************************************
Electronic Mail is not secure, may not be read every day, and should not be used for urgent or sensitive issues