[BioC] accessing slots in expresso

Laurent Gautier laurent@cbs.dtu.dk
Fri, 7 Feb 2003 04:27:40 +0100


I also remembered to roll in CVS a vignette I had about importing .CEL data from
alternative sources. Check the vignette vim.pdf in affy (devel, v.1.1.7)
or in the section 'vignettes' of the website (should appear w/ the next automated build).

L.


On Thu, Feb 06, 2003 at 09:58:15PM -0500, Rafael A. Irizarry wrote:
> this is what you have to do.
> 
> in the exprs (intensity) slot of an AffyBatch put a matrix with as many
> columns as arrays and as many row as number of probes on the array. 
> in each column put the vector of intensities of each array. a row must
> represent a unique location on the array. no you need to create a hash
> table enviromnet with varialbles with names of the probesets. the variabes
> need to be 2-column matrix where the first column has the positions of the
> PMs of that probe set and the second row the MMs. for example the variable
> 
> 1000_at
> could be
> 
> 1 641
> 2 642
> 12 652
> etc..
> 
> now in the cdfName slot of the exprset put the name of this enviromnet.
> this should get you on your way. look at the example
> 
> data(Dilution) 
> 
> to get a better idea. you will that the environment (loaded through a
> library) hgu95av2cdf has the variables 1000_at, 1001_at, etc... with the
> positions.
> 
> 
> hope this helps,
> -r
> 
> 
> 
> On Tue, 4 Feb 2003, Tibor van Rooij wrote:
> 
> > Hi! I would very much like to use the MBEI method, however all my data is stored in a database. I can easily access data such as probename x,y positions, pm value mm value etcetera. In the previous version of "affy" I could access slotnames of the objects such as names to put my names into and pm to put my pm values in and get the sytem to calculate the normalization ( so I did not need the original .CEL files)
> > 
> > I have been looking at the descriptions of the new classes and reading the manual but I cannot find out how to this in this new version.
> > 
> > Has anybody else attempted this?
> > 
> > Please let me know.
> > 
> > Tibor van Rooij 
> > Bioinformatician 
> > McGill University and
> > Genome Quebec Innovation Centre
> > 740 Dr. Penfield Avenue, Rm 6212
> > Montreal, Quebec H3A 1A4
> > Tel: (514) 398-2497
> > 
> > 
> > 	[[alternate HTML version deleted]]
> > 
> > _______________________________________________
> > Bioconductor mailing list
> > Bioconductor@stat.math.ethz.ch
> > http://www.stat.math.ethz.ch/mailman/listinfo/bioconductor
> >
> 
> _______________________________________________
> Bioconductor mailing list
> Bioconductor@stat.math.ethz.ch
> http://www.stat.math.ethz.ch/mailman/listinfo/bioconductor