problem with vsn and affy data...
2
0
Entering edit mode
@oliver-hartmann-141
Last seen 7.1 years ago
Dear list members, I am running into the error message Error: cannot allocate vector of size 22776 Kb. This is on a WinNT computer with 1 GB RAM, 1.8 GB RAM allocated to R (~1GB was in use bevor the error occured). I analyzed 24 affy Hg_u95a chips with R 1.6.1, Bioconductor 1.1, affy 1.1.1. I have analyzed 34 Affy chips (rma method) with Bioconductor 1.0 bevor (using a total of ~1.5 GB RAM) as well as 8 Affy Hg_u95a chips with R 1.6.1, Bioconductor 1.1, affy 1.1.1. (vsn and rma). The R script I used was > normalize.AffyBatch.methods <- c(normalize.AffyBatch.methods, "vsn") > f <- express(data, bg="subtractmm", normalize.method="vsn", summary.stat=avdiff) and I got: Normalizing Data Called vsn on 201807x24 matrix, with lts.quantile=0.75 and niter=10. This may take a while. This broke off after iter 7. Any ideas? Thanks a lot -oliver hartmann- -- Oliver Hartmann, Institute of Medical Biometry and Epidemiology Philipps-University Marburg, Bunsenstr. 3, D-35037 Marburg phone +49(0)6421 28 66514, fax +49(0)6421 28 68921
affy vsn affy vsn • 764 views
ADD COMMENT
0
Entering edit mode
@stephen-henderson-71
Last seen 4.5 years ago
Sorry. What is the vsn method you're talking about? I don't see this in my normalize.AffyBatch.methods Stephen Henderson -----Original Message----- From: Wolfgang Huber [mailto:w.huber@dkfz-heidelberg.de] Sent: Wednesday, December 18, 2002 3:11 PM To: Oliver Hartmann; bioconductor Subject: RE: [BioC] problem with vsn and affy data... Hi Oliver, the 201807 x 24 data matrix uses ca. 38 MB of memory (assuming 8 byte per double). Even if this data is copied back and forth a couple of times the total memory usage should not overrun 1 GB. So it seems to me more like a bug -either in vsn or in R- than just the "normal" excessive memory consumption. Have you tried with gc() before calling express() ? Otherwise, if you email me the necessary script and data set or put it on a ftp/http server I could offer you to try to reproduce the error here, both on windows or on a bigger Unix machine, and see what's causing it. Also, you can try different values for the parameters of vsn than the default (niter=10, lts.quantile=0.75). niter=5 and lts.quantile=0.5 should be sufficient for affy data and save some computing time. There was a problem with the affy::expresso/vsn interface shortly before the release in November, so can you please make sure you use the most recent version 1.0.1. of vsn? Best regards Wolfgang Huber Division of Molecular Genome Analysis (Dep. Poustka) German Cancer Research Center (DKFZ) Im Neuenheimer Feld 580 69120 Heidelberg, Germany w.huber@dkfz.de http://www.dkfz.de/abt0840/whuber Tel +49-6221-424709 Fax +49-6221-42524709 -----Original Message----- From: bioconductor-admin@stat.math.ethz.ch [mailto:bioconductor-admin@stat.math.ethz.ch]On Behalf Of Oliver Hartmann Sent: Wednesday, December 18, 2002 9:01 AM To: bioconductor@stat.math.ethz.ch Subject: [BioC] problem with vsn and affy data... Dear list members, I am running into the error message Error: cannot allocate vector of size 22776 Kb. This is on a WinNT computer with 1 GB RAM, 1.8 GB RAM allocated to R (~1GB was in use bevor the error occured). I analyzed 24 affy Hg_u95a chips with R 1.6.1, Bioconductor 1.1, affy 1.1.1. I have analyzed 34 Affy chips (rma method) with Bioconductor 1.0 bevor (using a total of ~1.5 GB RAM) as well as 8 Affy Hg_u95a chips with R 1.6.1, Bioconductor 1.1, affy 1.1.1. (vsn and rma). The R script I used was > normalize.AffyBatch.methods <- c(normalize.AffyBatch.methods, "vsn") > f <- express(data, bg="subtractmm", normalize.method="vsn", summary.stat=avdiff) and I got: Normalizing Data Called vsn on 201807x24 matrix, with lts.quantile=0.75 and niter=10. This may take a while. This broke off after iter 7. Any ideas? Thanks a lot -oliver hartmann- -- Oliver Hartmann, Institute of Medical Biometry and Epidemiology Philipps-University Marburg, Bunsenstr. 3, D-35037 Marburg phone +49(0)6421 28 66514, fax +49(0)6421 28 68921 _______________________________________________ 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 ********************************************************************** This email and any files transmitted with it are confidential an ... [[dropped]]
ADD COMMENT
0
Entering edit mode
Hi Stephen, vsn is another package on bioconductor.org that estimates for each array a normalization factor and an offset. It then performs a (kind of) log- linear hybrid transformation of the data such that the resulting differences are "log-ratios" with a constant variance along the whole intensity range. Equivalently, you can consider them z-scores of log-ratios, i.e. raw log-ratios divided by their variance. See below web site (Publications) for more details. The method can be applied to both cDNA and affy probe intensities, hence it's a separate package. It can be added to the normalize.AffyBatch.methods list through the mechanism described in the affy documentation, or also in the vsn vignette. The implementation is somewhat slow at this point. Best regards - Wolfgang Huber Division of Molecular Genome Analysis (Dep. Poustka) German Cancer Research Center (DKFZ) Im Neuenheimer Feld 580 69120 Heidelberg, Germany w.huber@dkfz.de http://www.dkfz.de/abt0840/whuber Tel +49-6221-424709 Fax +49-6221-42524709 -----Original Message----- From: Stephen Henderson [mailto:s.henderson@ucl.ac.uk] Sent: Wednesday, December 18, 2002 5:06 PM To: 'Wolfgang Huber'; Oliver Hartmann; bioconductor Subject: RE: [BioC] problem with vsn and affy data... Sorry. What is the vsn method you're talking about? I don't see this in my normalize.AffyBatch.methods Stephen Henderson
ADD REPLY
0
Entering edit mode
@wolfgang-huber-3550
Last seen 11 weeks ago
EMBL European Molecular Biology Laborat…
Hi Oliver, the 201807 x 24 data matrix uses ca. 38 MB of memory (assuming 8 byte per double). Even if this data is copied back and forth a couple of times the total memory usage should not overrun 1 GB. So it seems to me more like a bug -either in vsn or in R- than just the "normal" excessive memory consumption. Have you tried with gc() before calling express() ? Otherwise, if you email me the necessary script and data set or put it on a ftp/http server I could offer you to try to reproduce the error here, both on windows or on a bigger Unix machine, and see what's causing it. Also, you can try different values for the parameters of vsn than the default (niter=10, lts.quantile=0.75). niter=5 and lts.quantile=0.5 should be sufficient for affy data and save some computing time. There was a problem with the affy::expresso/vsn interface shortly before the release in November, so can you please make sure you use the most recent version 1.0.1. of vsn? Best regards Wolfgang Huber Division of Molecular Genome Analysis (Dep. Poustka) German Cancer Research Center (DKFZ) Im Neuenheimer Feld 580 69120 Heidelberg, Germany w.huber@dkfz.de http://www.dkfz.de/abt0840/whuber Tel +49-6221-424709 Fax +49-6221-42524709 -----Original Message----- From: bioconductor-admin@stat.math.ethz.ch [mailto:bioconductor-admin@stat.math.ethz.ch]On Behalf Of Oliver Hartmann Sent: Wednesday, December 18, 2002 9:01 AM To: bioconductor@stat.math.ethz.ch Subject: [BioC] problem with vsn and affy data... Dear list members, I am running into the error message Error: cannot allocate vector of size 22776 Kb. This is on a WinNT computer with 1 GB RAM, 1.8 GB RAM allocated to R (~1GB was in use bevor the error occured). I analyzed 24 affy Hg_u95a chips with R 1.6.1, Bioconductor 1.1, affy 1.1.1. I have analyzed 34 Affy chips (rma method) with Bioconductor 1.0 bevor (using a total of ~1.5 GB RAM) as well as 8 Affy Hg_u95a chips with R 1.6.1, Bioconductor 1.1, affy 1.1.1. (vsn and rma). The R script I used was > normalize.AffyBatch.methods <- c(normalize.AffyBatch.methods, "vsn") > f <- express(data, bg="subtractmm", normalize.method="vsn", summary.stat=avdiff) and I got: Normalizing Data Called vsn on 201807x24 matrix, with lts.quantile=0.75 and niter=10. This may take a while. This broke off after iter 7. Any ideas? Thanks a lot -oliver hartmann- -- Oliver Hartmann, Institute of Medical Biometry and Epidemiology Philipps-University Marburg, Bunsenstr. 3, D-35037 Marburg phone +49(0)6421 28 66514, fax +49(0)6421 28 68921 _______________________________________________ Bioconductor mailing list Bioconductor@stat.math.ethz.ch http://www.stat.math.ethz.ch/mailman/listinfo/bioconductor
ADD COMMENT

Login before adding your answer.

Traffic: 299 users visited in the last hour
Help About
FAQ
Access RSS
API
Stats

Use of this site constitutes acceptance of our User Agreement and Privacy Policy.

Powered by the version 2.3.6