[BioC] Using proportions...

[Wolfgang Huber]

On 11/17/11 4:58 AM, Jim Silverton wrote:
> Hello,
> Just wanted to find out why in a RNA-seq  with treatment and control, no
> one does the two sample proportion comparison for two features. The total
> reads are large so it should work asymptotically.
>

Dear Jim

lots of people do, but generally the answer that you get is not what one 
is looking for, because such an analysis tests against (and too 
frequenly rejects) an uninteresting null hypothesis. Have, for instance, 
a look at http://genomebiology.com/2010/11/10/R106 .

Best wishes
	Wolfgang



Wolfgang Huber
EMBL
http://www.embl.de/research/units/genome_biology/huber