the grep is retrieving some control samples out of the 278 columns in the matrix I used. if you have mapped some symbols which you wish to extract onto the corresponding probes (in the vector 'some.probes') and there is no data to plot, that is a different issue. I suspect you do not have any columns labeled as being from CD34+ samples, thus no data was returned... sorry about leaving that in the example. try some.probes <- mget( c('MLL','TET1'), hgu133plus2ALIAS2PROBE ) named <- data.matrix(my.exprs[ some.probes, ]) names(named) <- mget(rownames(named), hgu133plus2SYMBOL) heatmap(named, main='a heatmap') and see if that doesn't produce the desired result (at least for MLL and TET1 ;-)) On Tue, Feb 15, 2011 at 4:53 PM, Pete Shepard <peter.shepard@gmail.com>wrote: > Hi Tim, > > That worked at least in getting the names in the list, although I don't > understand what this grep section is doing? I just want the names in > some.probes? > > > named <- data.matrix(my.exprs[ some.probes, > grep('CD34', names(my.exprs)) ]) > > Sadly: > > > heatmap(named, main='a heatmap') > > gives the same error as before: > > 'x' must have at least 2 rows and 2 columns > > > > > On Tue, Feb 15, 2011 at 3:41 PM, Tim Triche, Jr. <ttriche@usc.edu> wrote: > >> alternative with probe names as row names (now I remember why it's better >> that way): >> >> > some.probes <- unlist(mget(c('ASXL1','WIT1','DAPK1'), >> hgu133plus2ALIAS2PROBE)) >> > named <- data.matrix(my.exprs[ some.probes, >> grep('CD34', names(my.exprs)) ]) >> > rownames(named) <- make.names( unlist(mget(some.probes, >> hgu133plus2SYMBOL)) ) >> > heatmap(named, main='a heatmap') >> >> >> >> >> On Tue, Feb 15, 2011 at 3:10 PM, Pete Shepard <peter.shepard@gmail.com>wrote: >> >>> Hi Tim, >>> >>> Thanks for your response. >>> >>> rownames(dat.n) <- make.names(mget(rownames(dat.n), hgu133a2SYMBOL, >>> ifnotfound=NA)) >>> >>> and >>> >>> rownames(dat.n) <- mget(rownames(dat.n), hgu133a2SYMBOL, ifnotfound=NA) >>> >>> seems to give me the same output. >>> >>> dat.plotme <- dat.n[ "NPEPL1",] >>> >>> using this command, I can only enter one gene symbol. >>> >>> If I run: >>> >>> heatmap(dat.plotme) >>> >>> I get the folllowing output: >>> >>> Error in heatmap(dat.plotme) : 'x' must be a numeric matrix >>> >>> Confused here >>> >>> On Tue, Feb 15, 2011 at 1:58 PM, Tim Triche <tim.triche@gmail.com>wrote: >>> >>>> library(hgu133plus2.db) >>>> rownames(dat.n) <- mget(rownames(dat.n), hgu133plus2SYMBOL, >>>> ifnotfound=NA) >>>> dat.plotme <- dat.n[ some.symbols, ] >>>> heatmap(dat.plotme) >>>> >>>> >>>> On Tue, Feb 15, 2011 at 1:43 PM, Pete Shepard <peter.shepard@gmail.com>wrote: >>>> >>>>> Dear List, >>>>> >>>>> I am attempting to visualize some affymetrix data from the hgu133a2.db >>>>> chip >>>>> using heatmap. So far, what I've done is as follows >>>>> >>>>> dat<-ReadAffy(filenames = "GSM448823.CEL", >>>>> "GSM448824.CEL","GSM448816.CEL", >>>>> "GSM448817.CEL") #puts .CEL files into a data frame >>>>> >>>>> dat2<-rma(dat) #normalizes data >>>>> >>>>> dat.n<-exprs(dat2) >>>>> >>>>> >>>>> What I would like to do next is for a subset of gene symbols, ~ 10, I >>>>> would >>>>> like to make a heatmap and display the gene symbols instead of the >>>>> probe >>>>> ids. >>>>> >>>>> So, I need to first get gene symbols from the probe ids and then select >>>>> only >>>>> a subset of these symbols from "dat.n" and make a heatmap of these >>>>> symbols. >>>>> >>>>> Can anyone help me regarding this problem? >>>>> >>>>> Thanks >>>>> >>>>> [[alternative HTML version deleted]] >>>>> >>>>> _______________________________________________ >>>>> Bioconductor mailing list >>>>> Bioconductor@r-project.org >>>>> https://stat.ethz.ch/mailman/listinfo/bioconductor >>>>> Search the archives: >>>>> http://news.gmane.org/gmane.science.biology.informatics.conductor >>>>> >>>> >>>> >>>> >>>> -- >>>> If people do not believe that mathematics is simple, it is only because >>>> they do not realize how complicated life is. >>>> John von Neumann<http: www-groups.dcs.st-="" and.ac.uk="" %7ehistory="" biographies="" von_neumann.html=""> >>>> >>>> >>> >> >> >> -- >> Pure reason shows that reason is not enough. >> >> Dennis V. Lindley<http: www.york.ac.uk="" depts="" maths="" histstat="" people="" lindley.gif=""> >> >> > -- Pure reason shows that reason is not enough. Dennis V. Lindley<http: www.york.ac.uk="" depts="" maths="" histstat="" people="" lindley.gif=""> [[alternative HTML version deleted]]

Hi, On Tue, Feb 15, 2011 at 7:53 PM, Pete Shepard <peter.shepard at="" gmail.com=""> wrote: > Hi Tim, > > That worked at least in getting the names in the list, although I don't > understand what this grep section is doing? Just look at what the function call grep returns: > I just want the names in some.probes? > > named <- data.matrix(my.exprs[ some.probes, > ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? grep('CD34', names(my.exprs)) ]) For example: R> wut <- grep('CD34', names(my.exprs)) R> head(wut) R> head(names(my.exprs)[wut]) Alternatively, you could always read the documentation found via ?grep > Sadly: > > heatmap(named, main='a heatmap') > > gives the same error as before: > > 'x' must have at least 2 rows and 2 columns Your previous error said that "x must be a numeric matrix", and now it's telling you that x needs at least two rows and two columns. The `x` that is being referred to is the `x` you are passing into the heatmap function, so you might want to take a look to see what `x` is. In this case, you are setting `named` to x, so what is the output of: R> is(named) R> dim(named) R> head(named) A little exploration of your data (or what you assume your data is) via the intermediary vars/steps you are creating will get you a long way. -- 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