[BioC] specifying contrast in edgeR

Ryan C. Thompson rct at thompsonclan.org
Tue Dec 11 04:09:12 CET 2012


glmLRT doesn't take a design argument. Just a coef or contrast. I 
believe the documentation for glmLRT indicates that the contrast takes 
precedence over the coef.

On Mon 10 Dec 2012 07:04:16 PM PST, Alpesh Querer wrote:
> Thanks Ryan,
>
> so lrt <- glmLRT(fit,contrast=c(1,-0.5-,-0.5),design) will override
> the default coef  and coef is no longer required to be specified?
>
> Regards,
> Al
>
> On Mon, Dec 10, 2012 at 8:50 PM, Ryan C. Thompson
> <rct at thompsonclan.org <mailto:rct at thompsonclan.org>> wrote:
>
>     glmFit doesn't take a contrast argument as far as I know. You only
>     specify the contrast in glmLRT.
>
>
>     On Mon 10 Dec 2012 06:47:40 PM PST, Alpesh Querer wrote:
>
>         Hi all,
>
>         I want to test the hypothesis Cond1-(Cond2+Cond3)/2 =0 in
>         glmFIT, do I need
>         to specify (1,-0.5,-0.5) as the contrast vector in glmFit?
>         Also should I include this contrast in glmLRT as well or it
>         automatically
>         uses the contrast used in the previous step?
>         Will the following commands give me what I want?
>
>         fit <- glmFit(dgelist_object, design,contrast=c(1,-0.5-,-0.__5))
>         lrt <- glmLRT(fit)
>
>         Thanks,
>         Al
>
>                 [[alternative HTML version deleted]]
>
>         _________________________________________________
>         Bioconductor mailing list
>         Bioconductor at r-project.org <mailto:Bioconductor at r-project.org>
>         https://stat.ethz.ch/mailman/__listinfo/bioconductor
>         <https://stat.ethz.ch/mailman/listinfo/bioconductor>
>         Search the archives:
>         http://news.gmane.org/gmane.__science.biology.informatics.__conductor
>         <http://news.gmane.org/gmane.science.biology.informatics.conductor>
>
>



More information about the Bioconductor mailing list