Richard Friedman friedman at cancercenter.columbia.edu
Wed Jan 25 15:46:11 CET 2006


	GCRMA works better than RMA. I generally recommend that
a B-value cutoff be determined by spot-checking by PCR.

I hope this helps,

On Jan 24, 2006, at 10:51 AM, cap2018 at columbia.edu wrote:

> I have a set of microarray experiments to which I have applied both
> the rma and gcrma preprocessing.  What I have read seemed to
> indicate that the gcrma is better, however I am having issues with
> p values resulting from my relevant comparison within the study.
> There are 2 experimental factors, brain region and line. There are 6
> chips in each group, 24 total (6 region1:line1, 6 region2.line1,
> etc).  I applied a 2-way ANOVA resulting in p values for region,
> line and their interaction.  When I make a frequency histogram of
> the results for line (the most important comparison) for RMA, the
> results look as expected with the largest number of pvalues close
> to zero. When I make the same histogram for my GCRMA results, the
> plot looks different, with the largest number of pvalues centered
> around .2.  I wanted to apply statistical methods to figure out the
> False Discovery Rate of the tests, but I'm not sure they are
> relevant to these GCRMA results.
> Any comments on GCRMA or FDR test would be helpful.
> Thanks
> Christine
> _______________________________________________
> Bioconductor mailing list
> Bioconductor at stat.math.ethz.ch
> https://stat.ethz.ch/mailman/listinfo/bioconductor
Richard A. Friedman, PhD
Associate Research Scientist
Herbert Irving Comprehensive Cancer Center
Oncoinformatics Core
Department of Biomedical Informatics
Box 95, Room 130BB or P&S 1-420C
Columbia University Medical Center
630 W. 168th St.
New York, NY 10032
(212)305-6901 (5-6901) (voice)
friedman at cancercenter.columbia.edu

"42 is the answer. Dylan got it wrong. 'Blowin'
in the wind' is not the answer. It isn't even
a number' " - Rose Friedman, age 9

More information about the Bioconductor mailing list