[BioC] simpleaffy finds QC problem, now what?

[Anand C.Patel]

I have 12 U74Av2 microarrays (and their B and C counterparts).  I used
SimpleAffy to make an AffyBatch of the raw data from the CEL files, and
normalized with MAS5.  On plotting QC, I discovered that 3 of the arrays were
outliers (see plot attached).  Looking at AffyRNADeg, the 3 arrays have a
different slope than the rest of the arrays (plot attached).

Questions are:
1.  Does the degradation difference explain the normalization problem?
2.  Is there a way to compensate for this?
3.  Could one normalize those separately and then somehow cross normalize the
two groups of arrays (in terms of degradation properties)?

I also normalized with GCRMA, but the QC tools don't work the same way (as
they won't given the nature of the tools and the normalization algorithms),
but they still don't look like the rest of the group (box plot attached).

Thanks,
Anand C. Patel, MD
Washington University School of Medicine
acpatel at usa.net

-------------- next part --------------
A non-text attachment was scrubbed...
Name: qc_plot.png
Type: image/png
Size: 33030 bytes
Desc: not available
Url : https://stat.ethz.ch/pipermail/bioconductor/attachments/20050328/bd165742/qc_plot.png
-------------- next part --------------
A non-text attachment was scrubbed...
Name: rna_deg.png
Type: image/png
Size: 23680 bytes
Desc: not available
Url : https://stat.ethz.ch/pipermail/bioconductor/attachments/20050328/bd165742/rna_deg.png
-------------- next part --------------
A non-text attachment was scrubbed...
Name: gcrma_boxplot.png
Type: image/png
Size: 15334 bytes
Desc: not available
Url : https://stat.ethz.ch/pipermail/bioconductor/attachments/20050328/bd165742/gcrma_boxplot.png