[BioC] Testing for significant genes with RMA

Eric emblal at uky.edu
Mon Dec 22 16:02:31 MET 2003

Hi, thanks for the reply. I don't think I said that clearly enough.
I believe the situation is closer to:


The p-values from one-way ANOVA on the different data sets have extremely 
poor correlation (R2 ~ .1). It is my understanding that a straight 
subtraction/addition would not alter variance and so the p-values would be 
exactly the same, while a division/multiplication would alter variance- 
which appears to be the case when I compare the results of justRMA output 
to those of the "log of the fold change of justRMA using a control group".

At 09:00 AM 12/22/2003 +0100, you wrote:
>Date: Fri, 19 Dec 2003 10:20:59 -0500
>From: Naomi Altman <naomi at stat.psu.edu>
>Subject: Re: [BioC] Testing for significant genes with RMA
>To: Eric <emblal at uky.edu>, bioconductor at stat.math.ethz.ch
>Message-ID: < at stat.psu.edu>
>Content-Type: text/plain; charset="us-ascii"; format=flowed
>If I understand your question, it doesn't matter.  (y-m)-(x-m)=y-x.

At 10:00 AM 12/19/2003, Eric wrote:
 >Which would be more appropriate to use as the initial data for a 'per gene
 >parametric test', the output of justRMA or the output of justRMA normalized
 >to the mean of a control group (log of fold change)?

Eric Blalock, PhD
Dept Pharmacology, UKMC
859 323-8033


The contents of this e-mail message and any attachments are confidential 
and are intended solely for addressee. The information may also be legally 
privileged. This transmission is sent in trust, for the sole purpose of 
delivery to the intended recipient. If you have received this transmission 
in error, any use, reproduction or dissemination of this transmission is 
strictly prohibited. If you are not the intended recipient, please 
immediately notify the sender by reply e-mail or at (859) 323-8033 and 
delete this message and its attachments, if any.
	[[alternative HTML version deleted]]

More information about the Bioconductor mailing list