[BioC] zero values and lib.size

[James W. MacDonald]

On 2/10/2014 12:06 PM, J [guest] wrote:
> I'm using the "classic" edgeR method and I was wondering if it is okay to add 0.5 to all of my tag counts before I do any processing of the data to make sure edgeR p-values and and logFC are not effected by any genes with 0 counts?  If this is okay, do I need to use these column sums as the new lib.size or since it would all be the same for every column would it be unnecessary?
>
> I'm referring to doing something like this…
>
> all_counts <- all_counts + 0.5

I suppose you could, but you are assuming that the problem of zero 
counts never occurred to the author of edgeR. Is that really a bet you 
want to make? And if you want to make that bet, why on earth would you 
use edgeR?

Luckily for all involved, this is a solved problem, and you even get to 
provide input if you don't like the defaults. See the prior.count 
argument to ?exactTest.

Best,

Jim
>
> then preceding with making the DGElist etc.
>
> Thanks
>
>   -- output of sessionInfo():
>
> R version 3.0.2 (2013-09-25)
> Platform: x86_64-apple-darwin10.8.0 (64-bit)
>
> locale:
> [1] en_US.UTF-8/en_US.UTF-8/en_US.UTF-8/C/en_US.UTF-8/en_US.UTF-8
>
> attached base packages:
> [1] splines   stats     graphics  grDevices utils     datasets  methods
> [8] base
>
> other attached packages:
> [1] edgeR_3.4.2   limma_3.18.11
>
> loaded via a namespace (and not attached):
> [1] tools_3.0.2
>
> --
> 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

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