[BioC] edgeR: summary of differentially expressed genes or tags

[Gordon K Smyth]

Dear KJ Lim,

When testing for multiple coefficients, the easiest way to learn the 
number of differentially expressed genes is by:

   FDR <- p.adjust(ltr$PValue, method="BH")
   sum(FDR < 0.05)

There is however no unambiguous way to break this down by up and down 
genes.  There is no unambiguous way to classify a gene as up or down with 
respect to multiple coefficients, because the coefficients may change in 
different directions for the same gene.

Best wishes
Gordon

> Date: Sat, 2 Jun 2012 18:19:32 +0300
> From: KJ Lim <jinkeanlim at gmail.com>
> To: Bioconductor mailing list <bioconductor at r-project.org>
> Subject: [BioC] edgeR: summary of differentially expressed genes or
> 	tags
>
> Dear edgeR community,
>
> Good day.
>
> I can learn summary of the up and down regulated genes/tags from
>  >summary(de <- decideTestsDGE(lrt))
> when the *coef *of *glmLRT*(the likelihood ratio test) is set to one degree
> of freedom.
>
> When the *coef* is set to i.e. 2:5; the decideTestsDGE doesn't work. It
> could be nice to see the summary of up and down regulated genes/tags when
> the *coef* is set i.e. 2:5.
>
> Thus, may I ask is there any method to learn the number of up and down
> regulated genes/tags when the *coef* of *glmLRT*(the likelihood ratio test) is
> set to all groups?
>
> Thank you very much for your guys time and help.
>
> Have a nice weekend.
>
> Best regards,
> KJ Lim


______________________________________________________________________
The information in this email is confidential and intend...{{dropped:4}}