[BioC] htseq and dexseq

[Andreia Fonseca]

Hi,

You are right these cases are the minority. My feelling was to leave these apart but I was affraid to be discarding something important.
Thanks for the tips.

Kind regards,

Andreia

Sent from my iPad

On 26/06/2013, at 22:56, Steve Lianoglou <lianoglou.steve at gene.com> wrote:

> Hi,
> 
> On Wed, Jun 26, 2013 at 2:00 PM, Andreia Fonseca
> <andreia.fonseca at gmail.com> wrote:
>> Thanks Steve,
>> 
>> The vignette looks very useful. ;y question was not related with counting
>> into two exons of same gene. With Htseq we get counts for exons that
>> overlap different genes, so it creates feature fusing two or more genes,
>> and I don't know how to handle this. should I look into the counts of the
>> exons of the single gene
>> , or of the counts of exons the gene fusions?
> 
> Remember that I've never used ht-seq, but I'll assume that it's easy
> enough to find the "bins" that result from an exonic region shared
> between two different genes.
> 
> The way I'd handle these special cases is that I wouldn't handle them
> at all and I'd likely prefer to remove those bins from my initial
> analysis -- I suspect the number of such cases would be relatively few
> when compared to the number of "kosher" bins.
> 
> HTH,
> -steve
> 
> --
> Steve Lianoglou
> Computational Biologist
> Bioinformatics and Computational Biology
> Genentech