[BioC] getGEO - getting the .CEL files from GEO

[Sean Davis]

On Thu, Mar 13, 2008 at 5:31 PM, Tim Smith <tim_smith_666 at yahoo.com> wrote:
> Hi,
>
>  Is there any way that I can get the raw .CEL files using getGEO in R? I don't want to use the SOFT files, since I don't know what kind of normalization has been done on them.
>
>  thanks!
>
>  PS: I have also checked the previous emails on this topic, and still can't make out what I need to do to get the files.

Hi, Tim.  I haven't added a helper to do this because the supplemental
files at GEO can contain ANYTHING.  However, if you know that you have
a set of samples that contain supplemental .cel files, the process is
trivial.  This page

http://www.ncbi.nlm.nih.gov/projects/geo/info/geo_paccess.html#FTP

shows you exactly where the files are.  You can use download.file() to
loop through the GSM # or just download 1 large file via GSE number.
I suggest using getGEO() to get the soft format that includes the
metadata and platform information.  Then, simply replace the exprs()
with the locally-processed data.  You might think that GEOquery could
do all of this for you, and for affy, it could, but for the 4000 other
microarray platforms in GEO, it really isn't possible to do so.

Hope that helps.

Sean