[BioC] [Fwd: Re: gene level on exon chip?]

[Michal Okoniewski]

Hi Paul.. .

There was long discussion when I was implementing it and with the help 
of my coleagues  we came up with the solution that
it is used there now - you can check the code internals of  
.splicing.index()
Only exonic (see: select.probewise()) probesets are taken into account, 
then by defaut a median is taken (may be changed to mean by a parameter) 
and then get
"unlogged" values and average them and then log back. Hopefully, it 
follows the splicing index as described in Gardina et al, BMC Genomics 
7, 2006

Cheers,
Michal


Paul Hammer wrote:
>  
>
> @ michal: you are absolutely right. i know only to well how hard it is 
> to interpret gene levels on exon chips. but i try to compare results 
> from exon chips with gene chips. by the way how are you calculating 
> the gene level within the function splicing.index()?
>
> cheers
> paul
>
>