When I last worked with LumiBatch objects, I found that the [, -9] method did not subset control probes data. I'm pasting a work-around I used - apologies to maintainers that I didn't think at the time to report this. lumi.subset <- function(lumibatch.obj,index){ ##-------------------------------------------------------------------- ------ ##-------------------------------------------------------------------- ------ ## a work-around for a shortcoming in the lumibatch object, which does ## not subset control data when subsetting the object. ##-------------------------------------------------------------------- ------ ##lumibatch.obj: a lumibatch object to be subsetted ##index: an integer or logical vector specifying which samples to extract ## from the lumibatch object ##-------------------------------------------------------------------- ------ ##-------------------------------------------------------------------- ------ if(!(class(index)=="integer" | class(index)=="logical")){ error("index must be of type integer or logical") } ##temporarily set warn=-1 to suppress the warning you get when ##indexing a lumibatch object, since this function is a workaround ##to that limitation. options(warn=-1) lumibatch.reduced <- lumibatch.obj[,index] options(warn=0) controls.all <- getControlData(lumibatch.obj) controls <- controls.all[,c(match(c("controlType","ProbeID",sampleNames(lumibatch. reduced)),colnames(controls.all)))] if(all.equal(sampleNames(lumibatch.reduced),colnames(controls)[-1:-2 ])){ lumibatch.reduced <- addControlData2lumi(controls,lumibatch.reduced) print("Control data was correctly subsetted and inserted into the new lumibatch object.") }else{ error("couldn't match control data with reduced lumibatch object") } return(lumibatch.reduced) } On Fri, Aug 1, 2014 at 10:42 AM, James W. MacDonald <jmacdon@uw.edu> wrote: > Hi Anastasia, > > LumiBatch objects are direct extensions of the ExpressionSet class, so can > be subsetted using the '[' operator: > > > smallLumi <- lumiBarnes[,-9] > p> pData(phenoData(smallLumi)) > sampleID label pctBlood pctPlacenta replicate > A01 100USA A01 100 0 A > A02 95US:5PA A02 95 5 A > A03 75US:25PA A03 75 25 A > A04 50US:50PA A04 50 50 A > A05 25US:75PA A05 25 75 A > A06 100PA A06 0 100 A > A07 PBMC A07 NA NA A > A08 Illumina Control A08 NA NA A > B02 95US:5PB B02 95 5 B > B03 75US:25PB B03 75 25 B > B04 50US:50PB B04 50 50 B > B05 25US:75PB B05 25 75 B > B06 100PB B06 0 100 B > B07 PaxGene B07 NA NA B > B08 Illumina Control B08 NA NA B > > So you can see I removed the B01 sample. > > Best, > > Jim > > > > > On 7/31/2014 6:09 PM, Kilo Morkowi wrote: > >> Hi, >> >> I wil be very grateful if you could advise me on how can I correctly >> remove an outlier chip from llumiBatch object. >> >> For instance, let's assume that B01 sample in lumiBarnes example is an >> outlier. I'd like to remove it before normalization. >> >> How can this be done? Is here a method for deleting a chip? >> >> Many thanks for your help, >> --- Anastasia >> >> ps. >> I have imported data with lumiR function, added control info, modified >> phenoData, and made QC and plots. >> >> _______________________________________________ >> 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 >> >> > -- > James W. MacDonald, M.S. > Biostatistician > University of Washington > Environmental and Occupational Health Sciences > 4225 Roosevelt Way NE, # 100 > Seattle WA 98105-6099 > > > _______________________________________________ > 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 > -- Levi Waldron Assistant Professor of Biostatistics City University of New York School of Public Health, Hunter College 2180 3rd Ave Rm 538 New York NY 10035-4003 phone: 212-396-7747 www.waldronlab.org [[alternative HTML version deleted]]