Dear List, I have downloaded 100 samples of Affymetrix U133 Plus 2.0 array (more than 50,000 probes) from the web. We read those files into R by using ReadAffy( ) and tried carrying out the normalization by using rma( ). However we have encountered the problem of short of memory. I use Windows 2000 on PC and the memory is 1 GB. Is there some code in R for reading the CEL files one by one and some code to read the data row by row for rma, which may help our available memory deal with 'read files into R' and 'rma' commands. Has anyone met this problem with this size of dataset before and if so what can we do to correct the situation (possibly by running R on a unix server?). Dr. Stephen Moore Senior Scientist ArraGen Ltd. Almac House Seagoe Industrial Estate Craigavon Co. Armagh BT63 5QD Northern Ireland E-mail:Stephen.Moore@arragen.com Phone: 02838 395751 Mobile: 07719 989348 Fax: 02838 398676

I think you should try justRMA. See: http://files.protsuggest.org/biocond/html/2189.html and related posts. Sean On Aug 2, 2004, at 6:06 AM, Moore, Stephen wrote: > Dear List, > > I have downloaded 100 samples of Affymetrix U133 Plus 2.0 array (more > than 50,000 probes) from the web. We read those files into R by using > ReadAffy( ) and tried carrying out the normalization by using rma( ). > However we have encountered the problem of short of memory. I use > Windows 2000 on PC and the memory is 1 GB. Is there some code in R for > reading the CEL files one by one and some code to read the data row by > row for rma, which may help our available memory deal with 'read files > into R' and 'rma' commands. Has anyone met this problem with this > size of dataset before and if so what can we do to correct the > situation (possibly by running R on a unix server?). > > Dr. Stephen Moore > Senior Scientist > ArraGen Ltd. > Almac House > Seagoe Industrial Estate > Craigavon > Co. Armagh > BT63 5QD > Northern Ireland > E-mail:Stephen.Moore@arragen.com > Phone: 02838 395751 > Mobile: 07719 989348 > Fax: 02838 398676 > > _______________________________________________ > Bioconductor mailing list > Bioconductor@stat.math.ethz.ch > https://www.stat.math.ethz.ch/mailman/listinfo/bioconductor

On Monday 02 August 2004 12:06, Moore, Stephen wrote: i use linux and this system has a best control on swap memory. it's not very fast b > Dear List, > > I have downloaded 100 samples of Affymetrix U133 Plus 2.0 array (more than > 50,000 probes) from the web. We read those files into R by using ReadAffy( > ) and tried carrying out the normalization by using rma( ). However we have > encountered the problem of short of memory. I use Windows 2000 on PC and > the memory is 1 GB. Is there some code in R for reading the CEL files one > by one and some code to read the data row by row for rma, which may help > our available memory deal with 'read files into R' and 'rma' commands. Has > anyone met this problem with this size of dataset before and if so what can > we do to correct the situation (possibly by running R on a unix server?). > > Dr. Stephen Moore > Senior Scientist > ArraGen Ltd. > Almac House > Seagoe Industrial Estate > Craigavon > Co. Armagh > BT63 5QD > Northern Ireland > E-mail:Stephen.Moore@arragen.com > Phone: 02838 395751 > Mobile: 07719 989348 > Fax: 02838 398676 > > _______________________________________________ > Bioconductor mailing list > Bioconductor@stat.math.ethz.ch > https://www.stat.math.ethz.ch/mailman/listinfo/bioconductor -- Dott. Alberto Goldoni Lab. Genetica Medica Ospedale S.Orsola-Malpighi via Malpighi n.9, BO

Go to http://www.liondb.org/biocond and search for "many RMA" or "large RMA". Similar question has been asked many times before. Some suggestions include a) just.rma b) allocate more memory to R by --max-mem-size when starting R c) increase memory - physically or use other machines d) my sub-sampling approach HTH, Adai. On Mon, 2004-08-02 at 11:06, Moore, Stephen wrote: > Dear List, > > I have downloaded 100 samples of Affymetrix U133 Plus 2.0 array (more than 50,000 probes) from the web. We read those files into R by using ReadAffy( ) and tried carrying out the normalization by using rma( ). However we have encountered the problem of short of memory. I use Windows 2000 on PC and the memory is 1 GB. Is there some code in R for reading the CEL files one by one and some code to read the data row by row for rma, which may help our available memory deal with 'read files into R' and 'rma' commands. Has anyone met this problem with this size of dataset before and if so what can we do to correct the situation (possibly by running R on a unix server?). > > Dr. Stephen Moore > Senior Scientist > ArraGen Ltd. > Almac House > Seagoe Industrial Estate > Craigavon > Co. Armagh > BT63 5QD > Northern Ireland > E-mail:Stephen.Moore@arragen.com > Phone: 02838 395751 > Mobile: 07719 989348 > Fax: 02838 398676 > > _______________________________________________ > Bioconductor mailing list > Bioconductor@stat.math.ethz.ch > https://www.stat.math.ethz.ch/mailman/listinfo/bioconductor >