[BioC] GCRMA with HG-U95A and HG-U95Av2

Nicolas STRANSKY Nicolas.Stransky at curie.fr
Tue Dec 5 19:01:10 CET 2006


Hi,

I'm trying to normalize HG-U95A and HG-U95Av2 chips with GCRMA.
I have tried two methods that were described on this list but have been
unsuccessful with both.

The first one was proposed by Zhijin Wu and adapted by Vincent Detours
(http://article.gmane.org/gmane.science.biology.informatics.conductor/2747/).
It involved combineAffyBatch. I could not get it to work, it always ends
up with the following error:

> affinity.info1=compute.affinities("hgu95a")
Computing affinities.Done.
>
affinity.info1=new("AffyBatch",exprs=exprs(affinity.info1),cdfName=cdfName(affinity.info1),nrow=640,ncol=640)
> affinity.info2=compute.affinities("hgu95av2")
Computing affinities.Done.
>
affinity.info2=new("AffyBatch",exprs=exprs(affinity.info2),cdfName=cdfName(affinity.info2),nrow=640,ncol=640)
> affinity.mix12=combineAffyBatch(list(affinity.info1,affinity.info2),
c("hgu95aprobe","hgu95av2probe"), "hgu95av12new")
package:hgu95aprobe     hgu95aprobe
package:hgu95av2probe   hgu95av2probe
197193 unique probes in common
Error in "dimnames<-.data.frame"(`*tmp*`, value = list(c("Batch 1: ",  :
        invalid 'dimnames' given for data frame


The second solution involves the packages created by Ben Bolstad
(http://bmbolstad.com/misc/mixtureCDF/MixtureCDF.html) and the sample R
code from http://bmbolstad.com/misc/mixtureCDF/mixtureCDFexample.R.

> My.version1<- ReadAffy(celfile.path="./A")
> My.version2<- ReadAffy(celfile.path="./Av2")
> my.merged <-
merge.mix(My.version1,My.version2,mergecdfName="HGU95A_V12mix")
> gcrma(my.merged)
Attaching package: 'hgu95av12mixcdf'
Adjusting for optical effect..............................Done.
Computing affinities[1] "Checking to see if your internet connection
works..."
Error in getProbePackage(probepackagename) :
        You do not have write access to /usr/local/lib/R/site-library
Please check your permissions or provide a different 'lib' parameter.


In order to correct this error, I have tried to apply so solution
described here:
http://article.gmane.org/gmane.science.biology.informatics.conductor/6012
But the package I create is not installable for an unknown reason.

# R CMD INSTALL /tmp/hgu95av12mixprobe_1.14.0.tar.gz
* Installing *source* package 'hgu95av12mixprobe' ...
** R
** data
** help
 >>> Building/Updating help pages for package 'hgu95av12mixprobe'
     Formats: text html latex example
  hgu95av12mixprobe                 text    html    latex   example
** building package indices ...
Error in load(zfile, envir = envir) : ReadItem: unknown type 101
Execution halted
ERROR: installing package indices failed
** Removing '/usr/local/lib/R/site-library/hgu95av12mixprobe'


So has anybody been successful for a GCMRA normalization with U95A and
U95Av2 chips ?

Thanks for any help,
-- 
Nicolas STRANSKY
Équipe Oncologie Moléculaire             http://www.curie.fr/equipe/301
Institut Curie - UMR 144 - CNRS                 Tel : +33 1 42 34 63 40
26, rue d'Ulm - 75248 Paris Cedex 5 - FRANCE    Fax : +33 1 42 34 63 49



More information about the Bioconductor mailing list