[BioC] Using limma with Agilent array data

[Gordon K Smyth]

> Date: Mon, 2 Jun 2008 15:26:20 -0700
> From: "Anh Tran" <popophobia at gmail.com>
> Subject: [BioC] Using limma with Agilent array data
> To: "bioconductor at stat.math.ethz.ch" <bioconductor at stat.math.ethz.ch>
>
> Hi all,
> We've been using Agilent Scanner and Feature Extraction from Agilent.
>
> I'm wondering if there's a way to import these data into limma.

Already answered by Sean Davis.

> Out of the
> three required input files (GAL file, Targets file, Spot-type file), we only
> have GAL file.

The GAL file is not required.

The targets and spottypes files are created by you, not by Agilent.

> FE gives us SHP file and a txt file witn compiled list of
> probes and its reading. We are doing 2 color microarray btw.
>
> If you know any detail, please help us.
>
> Also, is it possible to group a certain probes together as one entity (they
> are expected to have the same result)?

If they really are the same probe, then yes, using avereps().

If they are not exactly the same probe, then no.

> And is it possible to give a set of
> probes as normalizing set for limma (set ratio to 1-1)?

You need to read the help file

   help(normalizeWithinArrays)

There is no normalization method in limma for which "set ratio to 1-1" 
would seem relevant.

> I've been looking through the manual but could not find any reference about
> Agilent output file.

The User's Guide is an overall guide, not the complete manual.  For 
detailed options you need to read the help pages for individual functions.

Best wishes
Gordon

> Thanks all
>
> -- 
> Regards,
> Anh Tran