[BioC] edgeR: confusing BCV plot

[Gordon K Smyth]

> Date: Wed, 12 Sep 2012 13:43:25 +0000
> From: Natasha Sahgal <nsahgal at well.ox.ac.uk>
> To: "James W. MacDonald" <jmacdon at uw.edu>
> Cc: "'bioconductor at r-project.org'" <bioconductor at r-project.org>
> Subject: Re: [BioC] edgeR: confusing BCV plot
>
> Dear Jim,
>
> Regarding the BCV plots, what I did not understand was the strange profile (at least strange to me), and the low coefficients of BV.
> Based on some figures from the user guide, it appeared to be very different - an increase towards the higher logCPM.
> 1) I'm not sure how to interpret these and if it is a good thing or not? (perhaps I have misunderstood the concept of the BCV)

Suggests to me that there is something unusual about your data. 
Especially the J shape at very low counts, which I have not seen before 
for RNA-seq data.

> 2) How does this affect the differential expression, if at all?

Genes that have larger estimated dispersions are less likely to be judged 
to be differentially expressed.

Gordon

> Re the filtering, for some reason I was under the impression increasing the counts per million would reduce (if not remove) zero counts in all samples. I agree with what you say about half the samples being unconstrained.
>
> I had 3 filters here, just to see what the difference would be. I still need to figure out the best or optimal one to use.
>
>
> Many Thanks and Best Wishes,
> Natasha
>

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