Hi David, since we also had a similar problem (actually, we wanted to apply fitPLM to more than 400 Plus2 chips), I wrote a just-Version of fitPLM. This just-version works for up to about 100 Plus2 chips on an AMD Athlon XP 3000+ with 1 GB RAM. I have also implemented a function that allows to fit PLM models to several hundreds of Plus2 arrays by splitting up the problem in several chunks. (The background correction and the quantile normalization is done chipwise, afterwards the chips are split in several chunks such that the resulting PLM signals are the same as the ones that would have been returned by fitPLM if fitPLM could have been used.) For details, see Chapter 5 of https://eldorado.uni- dortmund.de/bitstream/2003/23306/1/diss_schwender.pdf If you are interested in any of these functions then let me know. But please note that these functions were written more than a year ago using the BioC 1.8 (??) version of affyPLM. Haven't used these function since then. So no guarantee that they also work using the BioC 2.1 version of affyPLM. (But should work if Ben Bolstad did not recode the C code and the internal affyPLM functions in fitPLM. One problem might be that the resulting object is an object of class exprSet and not ExpressionSet.) I was thinking about publishing these functions. But didn't do it because of the R packages BufferedMatrix and oligo, in which the idea of splitting the problem into several chunks is implemented in a more elegant way. Best, Holger -------- Original-Nachricht -------- > Datum: Fri, 5 Oct 2007 13:57:53 +0200 (CEST) > Von: darteta001 at ikasle.ehu.es > An: bioconductor at stat.math.ethz.ch > Betreff: [BioC] reading many affy U133 Plus_2.0 > Dear list, > > In order to normalise my data using fitPLM(), I need to create an > affybatch using ReadAffy(). This time I think justRMA() is not an > option I can use. > > Has anyone been successful at reading in a large group of Plus chips > (say 100) using ReadAffy() and the normalise them using fitPLM() on > a "normal" PC on Windows? What amount of memory is needed for it? > Any examples on how many chips you might have loaded with memory > details would be very appreciated. > > Regards, > > David > > _______________________________________________ > Bioconductor mailing list > Bioconductor at stat.math.ethz.ch > https://stat.ethz.ch/mailman/listinfo/bioconductor > Search the archives: > http://news.gmane.org/gmane.science.biology.informatics.conductor -- Psssst! Schon vom neuen GMX MultiMessenger geh?rt? Der kanns mit allen: http://www.gmx.net/de/go/multimessenger

David, You might have a look at the R package 'aroma.affymetrix' ... you should be able to read/process/analyze 100s of affy chips. I regularly process >50 exon arrays, which are far larger than U133s. I believe you can do it all in less than 500M RAM. Full details are at: http://groups.google.com/group/aroma-affymetrix Cheers, Mark On 05/10/2007, at 9:57 PM, <darteta001 at="" ikasle.ehu.es=""> <darteta001 at="" ikasle.ehu.es=""> wrote: > Dear list, > > In order to normalise my data using fitPLM(), I need to create an > affybatch using ReadAffy(). This time I think justRMA() is not an > option I can use. > > Has anyone been successful at reading in a large group of Plus chips > (say 100) using ReadAffy() and the normalise them using fitPLM() on > a "normal" PC on Windows? What amount of memory is needed for it? > Any examples on how many chips you might have loaded with memory > details would be very appreciated. > > Regards, > > David > > _______________________________________________ > Bioconductor mailing list > Bioconductor at stat.math.ethz.ch > https://stat.ethz.ch/mailman/listinfo/bioconductor > Search the archives: http://news.gmane.org/ > gmane.science.biology.informatics.conductor