[BioC] Detection p value cut off

[James W. MacDonald]

Hi Hajja,

On 6/22/2011 7:13 AM, khadeeja ismail wrote:
> Hi Everyone,
>
> I'm new to Bioconductor and also microarrays. Currently I'm working
> on Illumina Human Methylation 450k data. I have done the color
> adjustments and ssn normalization. Now I need to remove the probes
> failing a detection p value cut-off. Would be great if someone could
> tell me the best way to do this.

The only 'detection p value' that I know of is based on a comparison 
between the PM and MM probes on an Affy 3' biased chip. Which isn't what 
you have. So I don't know that there is such a thing for you.

In addition, depending on how your analysis will proceed, you might not 
really want to do much filtering at the outset. For instance, if you are 
planning to use limma to do your comparisons, you want to keep all but 
the most aberrant probes when fitting the model. This is because you are 
using all data to estimate the expected variance of a probe, so if you 
filter first you may well be biasing this estimate, which is not ideal.

Best,

Jim


>
> Thanks, Hajja
>
>
>
> [[alternative HTML version deleted]]
>
>
>
>
> _______________________________________________ Bioconductor mailing
> list Bioconductor at r-project.org
> https://stat.ethz.ch/mailman/listinfo/bioconductor Search the
> archives:
> http://news.gmane.org/gmane.science.biology.informatics.conductor

-- 
James W. MacDonald, M.S.
Biostatistician
Douglas Lab
University of Michigan
Department of Human Genetics
5912 Buhl
1241 E. Catherine St.
Ann Arbor MI 48109-5618
734-615-7826
**********************************************************
Electronic Mail is not secure, may not be read every day, and should not be used for urgent or sensitive issues