Avhena, I would recommend that you not use R for this application. For hierarchical clustering of large numbers of genes + visualization I would recommend you try the free MeV program that can handle 20,000 genes easily. David On Wed, 17 Mar 2010, avehna wrote: > Hi again, > > I'm using a very powerful computer (a Mac Pro) with a huge amount of memory > and it's also very fast, that's why I wondered why this error came out. I > think the problem is that I'm using R in 32 bit mode. How could I change it? > > Regarding the amount of genes: For each treatment by separate the amount of > genes is smaller, but since I'm interested in making a heatplot to visualize > the gene profile across all treatments I'm using the union of all sets. I'm > very open to new suggestions from you. > > Thank you so much for your help, > > Avhena > > On Wed, Mar 17, 2010 at 6:09 PM, michael watson (IAH-C) < > michael.watson at bbsrc.ac.uk> wrote: > >> The first answers is that you need to make sure you computer has enough >> memory, and that R can access that memory. See memory.size() and >> memory.limit() for answers. >> >> The second answer is: are you *sure* you want to cluster 13,777 genes? >> That's an awful lot to visualise and interpret..... >> ________________________________________ >> From: bioconductor-bounces at stat.math.ethz.ch [ >> bioconductor-bounces at stat.math.ethz.ch] On Behalf Of avehna [ >> avhena at gmail.com] >> Sent: 17 March 2010 21:59 >> To: bioconductor at stat.math.ethz.ch >> Subject: [BioC] hclust for 13000 genes >> >> Hi There: >> >> I'm trying to cluster 13777 genes that are differentially expressed across >> three different treatments but I'm getting the following error: >> >> ------------------------------------------------ >> R(32949) malloc: *** mmap(size=1518448640) failed (error code=12) >> *** error: can't allocate region >> *** set a breakpoint in malloc_error_break to debug >> Error: cannot allocate vector of size 1.4 Gb >> Execution halted >> ------------------------------------------------- >> >> I wonder how could I fix it in order to get the clusters and the heatmap >> with the expression profile. I would appreciate any suggestion from you!. >> >> Thank you so much for your help in advance, >> >> Avhena >> >> [[alternative HTML version deleted]] >> >> _______________________________________________ >> 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 >> > > [[alternative HTML version deleted]] > > _______________________________________________ > 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 >

Hi, On Wed, Mar 17, 2010 at 9:04 PM, avehna <avhena at="" gmail.com=""> wrote: > Hi again, > > I'm using a very powerful computer (a Mac Pro) with a huge amount of memory That's nice ... but how much memory is that, exactly? > and it's also very fast, that's why I wondered why this error came out. I > think the problem is that I'm using R in 32 bit mode. How could I change it? It depends on how you launch R ... are you using R.app? or are you using R from the command line? Assuming you are using R.app, you'll need to launch the 64bit version. You should have an R64.app, which might have come in the default download ... what version of R are you using? You can actually get all the version of R and R.app necessary from here, too: http://r.research.att.com/ Read that page, and download what you need. Specifically search for the term "64-bit" on that page. -steve -- Steve Lianoglou Graduate Student: Computational Systems Biology | Memorial Sloan-Kettering Cancer Center | Weill Medical College of Cornell University Contact Info: http://cbio.mskcc.org/~lianos/contact

Hi Avehna When I do this, I use limma to fit a model that includes every factor, and the resulting output has both a F statistic and an F statistic p-value. These can be used to identify genes that are changing over any of your experiments However, I often find that even if I use a very stringent multiple testing procedure, I still end up with 10000s of genes to worry about. In my experience, this is not only difficult to visualise, but also to interpret. In the past I have applied other filters, such as the magnitude of change over time; or perhaps the lab scientist who produced the data may have created a time course, but is specifically interested in particular time points, so I filter on those; or sometimes, I just take the top n most changing genes, whether n be 500, 1000, 2000 etc. Thanks Mick From: avehna [mailto:avhena@gmail.com] Sent: 18 March 2010 01:05 To: michael watson (IAH-C); Steve Lianoglou; bioconductor@stat.math.ethz.ch Subject: Re: [BioC] hclust for 13000 genes Hi again, I'm using a very powerful computer (a Mac Pro) with a huge amount of memory and it's also very fast, that's why I wondered why this error came out. I think the problem is that I'm using R in 32 bit mode. How could I change it? Regarding the amount of genes: For each treatment by separate the amount of genes is smaller, but since I'm interested in making a heatplot to visualize the gene profile across all treatments I'm using the union of all sets. I'm very open to new suggestions from you. Thank you so much for your help, Avhena On Wed, Mar 17, 2010 at 6:09 PM, michael watson (IAH-C) <michael.watson@bbsrc.ac.uk<mailto:michael.watson@bbsrc.ac.uk>> wrote: The first answers is that you need to make sure you computer has enough memory, and that R can access that memory. See memory.size() and memory.limit() for answers. The second answer is: are you *sure* you want to cluster 13,777 genes? That's an awful lot to visualise and interpret..... ________________________________________ From: bioconductor-bounces@stat.math.ethz.ch<mailto:bioconductor- bounces@stat.math.ethz.ch=""> [bioconductor- bounces@stat.math.ethz.ch<mailto:bioconductor- bounces@stat.math.ethz.ch="">] On Behalf Of avehna [avhena@gmail.com<mailto:avhena@gmail.com>] Sent: 17 March 2010 21:59 To: bioconductor@stat.math.ethz.ch<mailto:bioconductor@stat.math.ethz.ch> Subject: [BioC] hclust for 13000 genes Hi There: I'm trying to cluster 13777 genes that are differentially expressed across three different treatments but I'm getting the following error: ------------------------------------------------ R(32949) malloc: *** mmap(size=1518448640) failed (error code=12) *** error: can't allocate region *** set a breakpoint in malloc_error_break to debug Error: cannot allocate vector of size 1.4 Gb Execution halted ------------------------------------------------- I wonder how could I fix it in order to get the clusters and the heatmap with the expression profile. I would appreciate any suggestion from you!. Thank you so much for your help in advance, Avhena [[alternative HTML version deleted]] _______________________________________________ Bioconductor mailing list Bioconductor@stat.math.ethz.ch<mailto:bioconductor@stat.math.ethz.ch> https://stat.ethz.ch/mailman/listinfo/bioconductor Search the archives: http://news.gmane.org/gmane.science.biology.informatics.conductor [[alternative HTML version deleted]]