[BioC] Limma and post statistical filtering of genes

[Gordon K Smyth]

Dear Sally,

You are rather expected to try reading the help pages and documentation 
before asking for help on this mailing list, and you don't seem to have 
done this.

Brief descriptions of all the components produced by lmFit() and eBayes() 
are given in the help page

   help("MArrayLM-class")

A more mathematical and technical description of the components in given 
in Sections 10.1 and 10.2 of the Limma User's Guide.

Best wishes
Gordon

> Date: Tue, 3 Feb 2009 13:01:36 -0800
> From: "Sally" <sagoldes at shaw.ca>
> Subject: [BioC] Limma and post statistical filtering of genes
> To: <bioconductor at stat.math.ethz.ch>
> Cc: Sally <sagoldes at shaw.ca>
> Content-Type: text/plain
>
> Hello,
>
> I have used Limma for preprocessing and statistical analysis.  I want to 
> filter my gene list after the statistical analysis but before the FDR is 
> applied.  My analysis generates the table fit2.  The fit2 table has the 
> following headers:
>
> Coefficients
> df.residual
> sigma
> stdev.unscaled
> A mean
> S2.post
> t
> p.value
> lods
> F
> F.p.value
>
> Can you explain what the following mean: stdev.unscaled, lods, s2.post, 
> Amean, coefficients and F.p.value mean?
>
> If I wanted to filter out genes by variance what would be the best way 
> to go?
>
> Sally