Re: FDR default in edgeR

"James W. MacDonald" <[email protected]> Wed, 3 Sep 2014 10:56:17 -0400
Newsgroups gmane.science.biology.informatics.conductor
Message-ID <CAKO-U0pf3YHCy6zpVcgk1Kzy=WoDrrrnEkbsB4eh8GOzTxbrCA@mail.gmail.com>
Hi Son,

If I understand your question correctly, there is no default. In other
words, if you look at the api for topTags, you see:

topTags(object, n=10, adjust.method="BH", sort.by="PValue")

and there isn't anything there to threshold the results by FDR. However,
you can use decideTestsDGE:

decideTestsDGE(object, adjust.method="BH", p.value=0.05)

which will allow you to select an FDR (or other (un)adjusted p-value) on
which to threshold, or you can use the default of 0.05.

Best,

Jim




On Wed, Sep 3, 2014 at 12:46 AM, Son Pham <spham-xrR1t/[email protected]> wrote:

> Hi All,
> Could you let me know what is the default value (or recommended value) for
> thresholding the FDR in edgeR?
> Thanks,
> Son.
>
>         [[alternative HTML version deleted]]
>
> _______________________________________________
> Bioconductor mailing list
> [email protected]
> https://stat.ethz.ch/mailman/listinfo/bioconductor
> Search the archives:
> http://news.gmane.org/gmane.science.biology.informatics.conductor
>



-- 
James W. MacDonald, M.S.
Biostatistician
University of Washington
Environmental and Occupational Health Sciences
4225 Roosevelt Way NE, # 100
Seattle WA 98105-6099

	[[alternative HTML version deleted]]

_______________________________________________
Bioconductor mailing list
[email protected]
https://stat.ethz.ch/mailman/listinfo/bioconductor
Search the archives: http://news.gmane.org/gmane.science.biology.informatics.conductor