[BioC] working with ROOT and the xps problem
cstrato
cstrato at aon.at
Fri Mar 18 19:27:50 CET 2011
Dear Assa,
First, please note that you can find extensive examples how to create
root scheme files in "xps/examples/script4schemes.R". For the miRNA-1_0
array it has the following information:
# directories for Affymetrix library and annotation files
libdir <- "/Volumes/GigaDrive/Affy/libraryfiles"
anndir <- "/Volumes/GigaDrive/Affy/Annotation"
scmdir <- "/Volumes/GigaDrive/CRAN/Workspaces/Schemes"
# miRNA-1_0:
# note: you need to rename "miRNA-1_0.probe_list.20081203.txt" to
"miRNA-1_0.probe.tab"
scheme.mirna10 <- import.expr.scheme("mirna10", filedir =
file.path(scmdir, "na31"),
schemefile = file.path(libdir, "miRNA-1_0.CDF"),
probefile = file.path(libdir, "miRNA-1_0.probe.tab"))
(You can ignore the warning messages you get.)
Once you have created file "mirna10.root" you can always load it with:
scheme.mirna10 <- root.scheme(paste(scmdir,"mirna10.root",sep="/"))
Regarding the new miRNA-2_0 array you need the corresponding
"miRNA-2_0.probe.tab" file. Since I could not find this file or the
corresponding "miRNA-2_0.probe_list.txt" file on the Affymetrix web site
you need to ask Affymetrix from where to obtain this file.
Please note that "miRNA-2_0.psi" is not a substitute for a probe.tab
file since it does not contain the necessary information.
In the future please provide always your sessionInfo().
Best regards
Christian
_._._._._._._._._._._._._._._._._._
C.h.r.i.s.t.i.a.n S.t.r.a.t.o.w.a
V.i.e.n.n.a A.u.s.t.r.i.a
e.m.a.i.l: cstrato at aon.at
_._._._._._._._._._._._._._._._._._
On 3/18/11 4:53 PM, Assa Yeroslaviz wrote:
> Hello bioc user,
>
> I have found a new package for the quality analysis and preprocessing of
> microarray data - xps.
> Unfortunately it is not so easy to use.
> I am working with the miRNA microarray from affymetrix.
>
> I tried to work according to the xps vignette and everything went just fine,
> up to the point where I wanted to analyze my own data...
>
> First I couldn't create a root file, as there is no explanation as to how to
> do it. I tried the command
>
> miRNA.test<- import.expr.scheme("mirna",filedir=getwd(),
>
>
> schemefile="miRNA-1_0_2Xgain.CDF",
> probefile="miRNA-1_0_2Xgain_probe.tab",
> annotfile=miRNA-1_0.annotations.20090219_annot.csv")
>
>
> but all I am getting is this message:
>
> ...
> Warning: probeset<cre-miR1157-star_st> has NumAtoms=4 and NumCells=4
> Warning: probeset<cre-miR1162-star_st> has NumAtoms=4 and NumCells=4
>
> Warning: probeset<cre-miR1165-star_st> has NumAtoms=4 and NumCells=4
> Warning: probeset<cre-miR1169-star_st> has NumAtoms=4 and NumCells=4
> Warning: probeset<mcmv-miR-m01-2-star_st> has NumAtoms=4 and NumCells=4
>
> Warning: probeset<mcmv-miR-m01-3-star_st> has NumAtoms=4 and NumCells=4
> Warning: probeset<mcmv-miR-m01-4-star_st> has NumAtoms=4 and NumCells=4
> Warning: probeset<mcmv-miR-M23-2-star_st> has NumAtoms=4 and NumCells=4
>
> Warning: probeset<mcmv-miR-m88-1-star_st> has NumAtoms=4 and NumCells=4
> Warning: probeset<mcmv-miR-m108-1-star_st> has NumAtoms=4 and NumCells=4
> Warning: probeset<mml-miR-1224-star_st> has NumAtoms=4 and NumCells=4
>
> Warning: probeset<hsa-miR-1226-star_st> has NumAtoms=4 and NumCells=4
> Warning: probeset<hsa-miR-1228-star_st> has NumAtoms=4 and NumCells=4
> Warning: probeset<hsa-miR-664-star_st> has NumAtoms=4 and NumCells=4
>
> Warning: probeset<mdv1-miR-M9-star_st> has NumAtoms=4 and NumCells=4
> Warning: probeset<cfa-miR-219-star_st> has NumAtoms=4 and NumCells=4
>
> <50001> records imported...
> <52900> records imported...Finished
> PM/MM statistics:
> 6703 cells with minimum number of PM/MM pairs: 2
> 78 cells with maximum number of PM/MM pairs: 47
>
>
> New dataset<miRNA-1_0_2Xgain> is added to Content...
> Importing</home/AYeroslaviz/Jenny_MicroRNA/miRNA-1_0_2Xgain_probe.tab>
> as<miRNA-1_0_2Xgain.prb>...
> Error: Header column must begin with<Probe Set Name> or<Probe Set ID>
>
> Error: Header line of file<miRNA-1_0_2Xgain_probe.tab> is not correct
> ...
> the file I used here is the *.PSI file from the annotation package of
> the affymetrix miRNA-1_0_2Xgain (to be used with FlashTag HSR). I just
> rename the end from PSI into tab. I didn't find any other file
>
>
> Does someone know where I can download the probe data for this array type?*
>
> *
> ...
> Importing</home/AYeroslaviz/Jenny_MicroRNA/miRNA-1_0.annotations.20090219_annot.csv>
> as<miRNA-1_0_2Xgain.ann>...
> Warning: Number of lines read<7816> is greater than number of genes<7815>
>
>
> Error: Wrong number of separators in line:
> 14q-0_st miRNA_v1 Homo sapiens Oct 21,
> 2008 CDBox =HYPERLINK(http://www-snorna.biotoul.fr/plus.php?id=14q(0) LINK
> to snoRNABase:14q(0)) 14q(0) chr14:100434010-100434086
> (+) 77 TGGACCAATGATGAGACAGTGTTTATGAACAAAAGATCATGATTAATCCAGTTCTGCACAAAACACTGAGGTCCATT
>
> Error when reading file<miRNA-1_0.annotations.20090219_annot.csv>.
>
> Apparently the annotation file is also not good enough. Somehow it
> count more lines than probes (?).
>
>
> I used here the annotation file for the newer chip of the miRNA
> (2.0),, which is probably the wrong one, but I just wanted to get it
> first to work, before I start tweaking with it.
>
> Error in import.expr.scheme("mirna", filedir = getwd(), schemefile =
> "miRNA-1_0_2Xgain.CDF", :
>
>
> error in function 'ImportExprSchemes'
>
>
> I am not exactly sure where I can find the right files for this package.
> A Google search found this site: http://microarray.bu.edu/~sshen/xps/, where
> complete root files were set to use.
>
> I download the mirna.root file and tried to excute it, but it didn't work
> either
>
> I just keep getting the error message:
>
>> mirna.scheme<- root.scheme(rootfile="mirna10.root")
> Warning in<TFile::ReadStreamerInfo>: mirna10.root: not a TStreamerInfo object
> Warning in<TFile::ReadStreamerInfo>: mirna10.root: not a TStreamerInfo object
> Warning in<TFile::ReadStreamerInfo>: mirna10.root: not a TStreamerInfo object
> Warning in<TFile::ReadStreamerInfo>: mirna10.root: not a TStreamerInfo object
>
>
> It will be nice if anyone has already done something like that and can help
> me here a little
>
> Thanks
> Assa
>
> [[alternative HTML version deleted]]
>
> _______________________________________________
> Bioconductor mailing list
> Bioconductor at r-project.org
> https://stat.ethz.ch/mailman/listinfo/bioconductor
> Search the archives: http://news.gmane.org/gmane.science.biology.informatics.conductor
>
More information about the Bioconductor
mailing list