[BioC] rma for different conditions

James W. MacDonald jmacdon at med.umich.edu
Tue May 22 15:50:36 CEST 2007

Hi Max,

MaximilianOtto at gmx.at wrote:
> Dear list, My question may be quite basic and it concerns the rma
> algorithm for affy data. As the idea behind rma is to take into
> account the between array variance for a probe set to calculate an
> expression value, I'm not sure if I can run rma on a set of cel files
> representing different conditions - for example  have 2 conditions,
> each replicated 3 times - 6 arrays total. should rma be run for all 6
> arrays or for each of three separately. in extreme I may have
> non-replicated experiments - use rma or not? Any suggestions are
> greatly appreciated! Thanks Max

You almost always want to run RMA on all chips at one time. Even when 
you are doing different conditions, most of the genes are not 
differentially expressed (at least that is the assumption most 
normalization schemes use). In addition, RMA is based on the idea that 
the pattern of probe intensities for a given probeset will be similar 
over different conditions, but the overall intensity may change. In 
other words, RMA is designed to account for similarities in the probe 
intensity pattern (likely non-Biological) while still detecting overall 
intensity differences.

Therefore, the expectation is that the data given to RMA will contain 
different conditions that you are planning on comparing.

That said, QA of the raw data is essential. RMA isn't magic, and if your 
data don't conform to the assumptions, the results you get will likely 
not be ideal. The affyQCReport package is one way to check the quality 
of your raw data.



> [[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.
Affymetrix and cDNA Microarray Core
University of Michigan Cancer Center
1500 E. Medical Center Drive
7410 CCGC
Ann Arbor MI 48109

Electronic Mail is not secure, may not be read every day, and should not be used for urgent or sensitive issues.

More information about the Bioconductor mailing list