[BioC] Separating gene filtering and multiple testing p-value adjustment steps

[Wolfgang Huber]

Dear Yoong

first, since you are using a recent feature, let's make sure we're all on the same page and please use the latest devel version:
http://www.bioconductor.org/packages/devel/bioc/html/DESeq2.html

How to turn off the independent filtering in the 'results' function is described in Section 3.6 of the vignette (http://www.bioconductor.org/packages/devel/bioc/vignettes/DESeq2/inst/doc/DESeq2.pdf ) and also in the manual page of the 'results' function.

You can manually combine your own filtering with your multiple-adjustment procedure of choice. The R code for doing this is rather trivial (even if the statistical concepts are perhaps not). See the vignette of the genefilter package "Diagnostics for independent filtering" for code examples and ideas.

The computation of size factors or normalisation factors (note that these are additional parameters that are kept with the data; the data themselves are not transformed) should be done before all differential expression analyses.

	Kind regards
	Wolfgang

Il giorno Dec 1, 2013, alle ore 3:21 am, FeiYian Yoon [guest] <guest at bioconductor.org> ha scritto:

> 
> Hi, 
> 
> In DESeq2, gene filtering and p-value adjustment for multiple testing are done under the same result function. I'd like to perform these two steps separately. To illustrate, I'd like to perform gene filtering for all my genes for all samples. But, for the multiple testing for differential expression, I'd like to do it with only a subset of samples for my comparisons of interest. How do I separate these two steps?
> 
> Also, I'd like to normalize my samples with endogenous genes that are stably expressed. Do I do it before p-value adjustment for multiple testing (differential expression)?
> 
> Thank you in advance!
> 
> Yoong
> 
> -- output of sessionInfo(): 
> 
> N/A
> 
> --
> Sent via the guest posting facility at bioconductor.org.
> 
> _______________________________________________
> Bioconductor mailing list
> Bioconductor at r-project.org
> https://stat.ethz.ch/mailman/listinfo/bioconductor
> Search the archives: http://news.gmane.org/gmane.science.biology.informatics.conductor