[BioC] how to design a model matrix and test by edgeR

[Gordon K Smyth]

On Thu, 26 Jul 2012, wang peter wrote:

> dear Dr Gordon:
>          thank you very much for your reply.
> for the first question, it should be cof=14.
> i made a mistake.
>
> and for the dispersion=dge$common.dispersion
> i make a mistake either.
>
> but thank you for youre corrrection.
>
> another question, if you donot mind.
> do i need estimate  the common dispersion
> before i estimate the tagwised dispersion

Try it, and you will have your answer.

If you run estimateGLMTagwiseDisp() without having previously estimated 
the common dispersion, it will give an error message to tell you that you 
need to run estimateGLMCommonDisp() first.

Gordon

> thank u
> shan
>

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