One problem here is that Illumina keeps changing their naming schemes both with new versions of the arrays and with new versions of BeadStudio. In the next release of the Illumina annotation files, I will include versions that are based on the ProbeID which I understand BeadStudio is now calling Array_Address. You can process files written out by the scanner software (BeadScanner) directly without using BeadStudio which use the ProbeID as the identifier so this should remain constant regardless of changes made to BeadStudio. These are the .csv and .txt files. The latter includes an entry for every probe type and is quite large. I recommend using the summarized version in the .csv format. I will also include a version based on the TargetID which may now be called ILMN_Gene. I believe the ProbeIDs are unique. Lynn Amon Jenny Drnevich wrote: > Hi Christian, > > I've been working with Illumina data myself and also came across this > error. The problem is that you're leaving the default value for > ProbeID to "TargetID" (see ?readBeadSummaryData for a description of > the arguments and their defaults), but you don't have a TargetID > column in your dataFile. Although the vignette and help page for > readBeadSummaryData both say the defaults are set for BeadStudio > Version 3 output, I've found this not to be true, at least for the > Mouse-6 array. In our case, there are a variety of different ID > numbers that can be output from BeadStudio, including both a PROBE_ID > and ProbeID (one's unique and one's not necessarily unique, I forget > which is which). You should look in your dataFile to see what is the > name of the column that contains the IDs and indicate that in the > call to readBeadSummaryData(). However, if the ID column contains > non-unique IDs, they are completely thrown out (a WARNING message > here would be nice!); you can check to see if the IDs came through by > looking at the rownames of exprs(BSData)[1:5,1:5] > > Interestingly, none of the IDs I can output from BeadStudio match the > probeIDs given in the IlluminaMouseV1 annotation package provided by > BioC. I'm going to investigate this further and if I figure out > what's going on, I'll send a message to the list. > > Cheers, > Jenny > > At 10:34 AM 3/18/2008, Christian Kohler wrote: > >> Dear Bioconductor-list, >> >> I recently started to go in for analysing Illumina (BeadChip) datasets >> with the help of the 'beadarray' package. >> As an Example R-script (within BeadSummaryExample) is provided, the >> introduction to this topic is quite straightforward. >> >> Until today, I used 'beadarray' in version 1.7.0 and the code snippets >> of the example script worked out fine. >> >> With 'beadarray' version 1.7.6. I experienced the following problem: >> >> >>> BSData <- readBeadSummaryData(dataFile, qcFile=qcFile, >>> >> sampleSheet=sampleSheet,skip=7, columns=list(exprs="AVG_Signal", >> se.exprs="BEAD_STDEV", >NoBeads="Avg_NBEADS"), >> qc.columns=list(exprs="AVG.Signal", se.exprs="SeqVAR"),qc.sep="," >> ,sep=",", qc.skip=7, annoPkg="illuminaHumanv1") >> >>> Error in `row.names<-.data.frame`(`*tmp*`, value = "ProbeID") : >>> invalid 'row.names' length >>> >> Now, I am just wondering if other 'beadarray'-users came across this error . >> >> >> >> >>> sessionInfo() >>> >> R version 2.6.2 (2008-02-08) >> x86_64-unknown-linux-gnu >> >> locale: >> C >> >> attached base packages: >> [1] tools stats graphics grDevices utils datasets methods >> [8] base >> >> other attached packages: >> [1] beadarray_1.7.6 affy_1.16.0 preprocessCore_1.0.0 >> [4] affyio_1.6.0 geneplotter_1.16.0 lattice_0.17-4 >> [7] annotate_1.16.0 xtable_1.5-1 AnnotationDbi_1.0.4 >> [10] RSQLite_0.6-3 DBI_0.2-3 Biobase_1.16.1 >> [13] limma_2.12.0 >> >> loaded via a namespace (and not attached): >> [1] KernSmooth_2.22-22 RColorBrewer_1.0-1 grid_2.6.2 rcompgen_0.1-17 >> [5] tcltk_2.6.2 >> >> >> >> >> >> >> Thanks a lot for any help/suggestions. >> >> Cheers >> Christian >> >> >> >> >> -- >> >> Christian Kohler >> Institute of Functional Genomics >> Computational Diagnostics >> University of Regensburg (BioPark I) >> D-93147 Regensburg (Germany) >> >> Tel. +49 941 943 5055 >> Fax +49 941 943 5020 >> christian.kohler at klinik.uni-regensburg.de >> >> _______________________________________________ >> 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 >> > > Jenny Drnevich, Ph.D. > > Functional Genomics Bioinformatics Specialist > W.M. Keck Center for Comparative and Functional Genomics > Roy J. Carver Biotechnology Center > University of Illinois, Urbana-Champaign > > 330 ERML > 1201 W. Gregory Dr. > Urbana, IL 61801 > USA > > ph: 217-244-7355 > fax: 217-265-5066 > e-mail: drnevich at uiuc.edu > > _______________________________________________ > 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 Mark and Jenny, thanks a lot for your detailed answers. After heeding your advice, I now understand :-) Best, Christian Mark Dunning wrote: > Hi Christian, > > This script was a bit outdated I'm afraid. The package was recently > changed so that it would use the ProbeID column in the input files to > give a unique row name in the resulting ExpressionSet. Unfortunately, > the file provided with the package did not have this column, hence R > gives an error when we try and assign a vector of length 0 to the > names of a matrix. > > If you add the argument ProbeID= "TargetID" to readBeadSummaryData it > should work. Thanks to Vince Carey for also pointing this error out > and the solution. > > There is also a problem in the code where we try and use BiomaRT to > get information about a set of probes. The vector illuids passed to > getBM is never defined. However, if you set to something appropriate > (maybe the rownames of the topTable output) then this command should > work fine also. > > I will update the script shortly with these changes. > > Hope this helps, > > Mark > > > > > On 18 Mar 2008, at 15:34, Christian Kohler wrote: > >> Dear Bioconductor-list, >> >> I recently started to go in for analysing Illumina (BeadChip) datasets >> with the help of the 'beadarray' package. >> As an Example R-script (within BeadSummaryExample) is provided, the >> introduction to this topic is quite straightforward. >> >> Until today, I used 'beadarray' in version 1.7.0 and the code snippets >> of the example script worked out fine. >> >> With 'beadarray' version 1.7.6. I experienced the following problem: >> >>> BSData <- readBeadSummaryData(dataFile, qcFile=qcFile, >> sampleSheet=sampleSheet,skip=7, columns=list(exprs="AVG_Signal", >> se.exprs="BEAD_STDEV", >NoBeads="Avg_NBEADS"), >> qc.columns=list(exprs="AVG.Signal", se.exprs="SeqVAR"),qc.sep="," >> ,sep=",", qc.skip=7, annoPkg="illuminaHumanv1") >>> Error in `row.names<-.data.frame`(`*tmp*`, value = "ProbeID") : >>> invalid 'row.names' length >> >> Now, I am just wondering if other 'beadarray'-users came across this >> error . >> >> >> >>> sessionInfo() >> R version 2.6.2 (2008-02-08) >> x86_64-unknown-linux-gnu >> >> locale: >> C >> >> attached base packages: >> [1] tools stats graphics grDevices utils datasets methods >> [8] base >> >> other attached packages: >> [1] beadarray_1.7.6 affy_1.16.0 preprocessCore_1.0.0 >> [4] affyio_1.6.0 geneplotter_1.16.0 lattice_0.17-4 >> [7] annotate_1.16.0 xtable_1.5-1 AnnotationDbi_1.0.4 >> [10] RSQLite_0.6-3 DBI_0.2-3 Biobase_1.16.1 >> [13] limma_2.12.0 >> >> loaded via a namespace (and not attached): >> [1] KernSmooth_2.22-22 RColorBrewer_1.0-1 grid_2.6.2 >> rcompgen_0.1-17 >> [5] tcltk_2.6.2 >> >> >> >> >> >> >> Thanks a lot for any help/suggestions. >> >> Cheers >> Christian >> >> >> >> >> -- >> Christian Kohler >> Institute of Functional Genomics >> Computational Diagnostics >> University of Regensburg (BioPark I) >> D-93147 Regensburg (Germany) >> >> Tel. +49 941 943 5055 >> Fax +49 941 943 5020 >> christian.kohler at klinik.uni-regensburg.de >> >> _______________________________________________ >> 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 > -- Christian Kohler Institute of Functional Genomics Computational Diagnostics University of Regensburg (BioPark I) D-93147 Regensburg (Germany) Tel. +49 941 943 5055 Fax +49 941 943 5020 christian.kohler at klinik.uni-regensburg.de