[BioC] Bug in gcrma_0.4.0?

[James MacDonald]

Running gcrma 0.4.0 on a WinXP box with R-1.7.1, I get odd results when
selecting the empirical bayes background correction (U133A chips). I
tried with and without triple.goal, and the results are the same.
Attached are some MA plots of the same two chips processed using rma and
gcrma using empirical bayes and mle background estimation.

Not sure if the mail server is stripping attachments due to our recent
experiences with sobig.F, so I also cc'ed to Rafael Irizarray (sorry for
the double email Rafael...).

Jim



James W. MacDonald
Affymetrix and cDNA Microarray Core
University of Michigan Cancer Center
1500 E. Medical Center Drive
7410 CCGC
Ann Arbor MI 48109
734-647-5623
-------------- next part --------------
A non-text attachment was scrubbed...
Name: MvA_gcrma_mle.png
Type: image/png
Size: 10503 bytes
Desc: not available
Url : https://www.stat.math.ethz.ch/pipermail/bioconductor/attachments/20030822/53a3190e/MvA_gcrma_mle-0001.png
-------------- next part --------------
A non-text attachment was scrubbed...
Name: MvA_no_triplegoal.png
Type: image/png
Size: 12389 bytes
Desc: not available
Url : https://www.stat.math.ethz.ch/pipermail/bioconductor/attachments/20030822/53a3190e/MvA_no_triplegoal-0001.png
-------------- next part --------------
A non-text attachment was scrubbed...
Name: MvA_rma.png
Type: image/png
Size: 10503 bytes
Desc: not available
Url : https://www.stat.math.ethz.ch/pipermail/bioconductor/attachments/20030822/53a3190e/MvA_rma-0001.png
-------------- next part --------------
A non-text attachment was scrubbed...
Name: MvA_with_triplegoal.png
Type: image/png
Size: 12971 bytes
Desc: not available
Url : https://www.stat.math.ethz.ch/pipermail/bioconductor/attachments/20030822/53a3190e/MvA_with_triplegoal-0001.png