[BioC] expression call

[Ido M. Tamir]

On Tuesday 22 August 2006 20:24, James W. MacDonald wrote:
> Hi Galina,
>
> Glazko, Galina wrote:
> > Dear list
> >
> >
> >
> > I need to select genes which are actually expressed in a set of Affy
> > chips.
> >
> > I've heard about 'expression call' thing, but I do not know exactly
> > whether it is relevant to my problem.
>
> I think you mean 'present call', which measures whether or not the
> intensities of PM probes for a particular probeset are significantly
> different from the corresponding MM probes. This is based on the idea
> that MM probes measure background binding, so if the PM probes are
> different, the transcript is assumed to be present (or expressed).
>
> You can get these values by using the mas5calls() function in the affy
> package. See ?mas5calls for more information.

Hi,

how would one get presence or absence calls with gcrma normalized data?
Is normalizing it with gcrma and then adding the mas5calls the most
sensible or only way to do it?

Or is there a know cutoff for gcrma normalized data where people
have measured with RT-PCR that a transcript is actually not expressed?

thank you very much,
Ido