[BioC] [Limma] M value in contrast test

James W. MacDonald jmacdon at med.umich.edu
Fri Nov 11 20:01:08 CET 2005


Marcelo Luiz de Laia wrote:
> Dear James W. MacDonald,
> 
> I send directly to your mail box because I am not sure if this is 
> suitable for the maillist. You are free to correct me.

It is suitable for the mail list, so I am responding there. I think 
there are enough people for whom English is a second language (depending 
on where you live, you may argue that this group includes the Americans 
;-D) that a clarification might be enlightening.

> 
> James W. MacDonald escreveu:
> 
>> Marcelo Luiz de Laia wrote:
>>  
>>
>>> I did a contrast in our analisis like this:
>>>
>>> (mutant treated-mutant control)-(wildtype treated-wildtype control)
>>>
>>>   
>>
>> In other words, you are looking for genes that respond differently to 
>> treatment in the wild type samples as compared to the mutant.
>>  
>>
> ops! A little confusion here, for me.
> 
> In limma users guide (26/04/2005 p.36)I found: "Which genes respond 
> differently in the mutant relative to the wild-type?
> 
>> cont.dif <- makeContrasts(
> 
> +     Dif6hr =(mu.6hr-mu.0hr)-(wt.6hr-wt.0hr),
> +     Dif24hr=(mu.24hr-mu.6hr)-(wt.24hr-wt.6hr),...)"
> 
> Your advice frase above is oposite to limma users guide. Not?

No, not really. The idea here is that we are looking for genes that 
react to treatment differently for one set of samples as compared to the 
other. Since microarrays are comparative assays, we are always comparing 
one set to the other.

For instance, if a gene is unaffected in mutant samples when treated, 
but the expression goes down in wild type, then this gene responds 
differently in the mutant (no change) relative to the wild type 
(down-regulation).

We can also say that this gene responds differently to treatment in the 
wild type samples (down-regulation) as compared to the mutant (no change).

Therefore, both phrases have equivalent meanings.

Does that help?

Jim



> 
> Thank you very much, again.
> 
> Marcelo
> 
> 


-- 
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



More information about the Bioconductor mailing list