[BioC] Affymetrix exon arrays?

[ltboots]

Hi Steven;
Here is a link to the exon array environment.

ftp://geneserver.mine.nu/affy/moex10stv1cdf.affy.v2.tar.gz

It is based on the affy exon probeset mappings. It can run on a machine with 
4Gb of memory, but can easily use 8-12Gb depending on what your trying to do. 
There are other probeset environments available for this platform at 
Bioconductor (refseq, ensembl, etc.) which are much more memory friendly.

I haven't seen much for alternative transcript structure packages at 
Bioconductor yet. I have been doing most of my analysis with mysql tables and 
perl scripts. They seem to handle the enormous dataset size better. Mysql 
works well with the multi-level probe->exon->transcript->gene mappings.

Jesse  
****************************************************************
Jesse Salisbury  Functional Genomics  University of Maine
email:  jesse.salisbury at umit.maine.edu
ph:  207-731-8318   web:  http://geneserver.mine.nu/
*****************************************************************

On Thursday 05 October 2006 15:27, Steven McKinney wrote:
> Hi Jesse and Seth,
>
> I need to start processing Affy Mouse Exon 1.0 ST
> data.  I'm interested in the annotated data package -
> did anything further get done about creating a data package
> from Jesse's work?  If not, are results still available
> from Jesse via FTP?
>
> We'd like to assess fold changes between mouse types
> at the exon and gene level.  I've parsed affy-related
> vignettes and R/BioC email trails, and don't see much
> discussion on techniques for mapping exon data to
> the gene level in R (saw one email response from Johannes
> Rainer, April 10 2006, on getting gene level data out of
> Affy's ExACT software, but we'd like to start with CEL
> data and do our own normalization etc).
>
> Our cluster is not operational yet, so I will start processing
> 4 or 5 CEL files on an Apple Mac PowerPC G5 with 4 processors
> and 4GB RAM.  I'm installing the latest R2.4.0/BioC release
> and will see whether exon data can be processed in this environment.
> Not sure whether I need 64 bit R or whether BioC packages
> are 64bit aware.
>
> If any one has done such analyses, and could share ideas
> and/or R code examples, it would be very much appreciated.
>
>
> Best regards
>
> Steven McKinney
>
> Statistician
> Molecular Oncology and Breast Cancer Program
> British Columbia Cancer Research Centre
>
> email: smckinney at bccrc.ca
> tel: 604-675-8000 x7561
>
> BCCRC
> Molecular Oncology
> 675 West 10th Ave, Floor 4
> Vancouver B.C.
> V5Z 1L3
>
> Canada

--