Dear All, I'm using XPS to analyse Human Exon Arrays. I tried DABG and MAS5 to compute the detection calls and I'm getting very different results (It seems to me that things gets scrambled at some point). here is the code I used: root.path <- "/home/thomas/QC.reporting/batch27/ROOT_files" root.filename <- "batch2.root" data.path <- "/home/thomas/QC.reporting/batch27/ROOT_files" data.filename="data_batch27.root" output.path <- "/home/thomas/QC.reporting/batch27/" root.data <- load.root.data(root.path, root.filename, data.path, data.filename) data.rma <- rma(root.data, filedir = root.path, filename = paste("tmpdt", "RMA", root.filename, sep="_"), tmpdir = paste(getwd(), "tmp", sep="/"), background="antigenomic", normalize = TRUE, option = "exon", exonlevel = "all", verbose=TRUE) call.dabg <- dabg.call(root.data, filedir = root.path, filename = paste("tmpdt", "RMA", root.filename, sep="_"), option = "exon", exonlevel = "all", verbose = TRUE) call.mas5 <- mas5.call(root.data, filedir = root.path, filename = paste("tmpdt", "MAS5", root.filename, sep="_"), option = "exon", exonlevel = "all", verbose = TRUE) The parameters for each detection call: > call.dabg@params $alpha1 [1] 0.04 $alpha2 [1] 0.06 > call.mas5@params $tau [1] 0.015 $alpha1 [1] 0.04 $alpha2 [1] 0.06 $ignore [1] TRUE The results I obtain (in matter of Present/Marginal/Absent): PivotTable.txt using DABG: Parameter W764.dab W766.dab PercentAbsent 88.6888 90.155 PercentMarginal 2.8019 2.4824 PercentPresent 8.50933 7.36258 PivotTable.txt: using MAS5 Parameter W764.dc5 W766.dc5 PercentAbsent 9.48876 9.89098 PercentMarginal 1.96661 1.96727 PercentPresent 88.5446 88.1418 The results obtained contradicts each other. Is there any reason for that? What am I doing wrong here? Thanks in advance for your answers. Thomas [[alternative HTML version deleted]]

Dear Thomas, For exon arrays Affymetrix has developed DABG as detection calls, so I would recommend to use call.dabg(). Although MAS5 calls work with exon arrays you need to be careful with the interpretation. If you use MAS5 calls with exon arrays then you should subtract the background first as follows: > call.mas5<- mas5.call(root.data,..., bgcorrect.option="correctbg",...) This will reduce the present calls to about 20-35% present calls. In addition, the help file ?mas5.call mentions: "The defaults for tau, alpha1 and alpha2 correspond to those in MAS5.0 for expression arrays. However, when using this function for exon or whole genome arrays, new values for alpha1 and alpha2 must be determined." This means that users may need to experiment with the settings, because I do not have an answer for this :-( One more note since I see the you use option="exon": The recommended and supported options are option="transcript" to get the expression values for the transcripts, and option="probeset" to get the expression values of the probesets as defined by Affymetrix for exon arrays. Although the additional option="exon" is also possible it is not recommended. 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 9/14/11 7:14 PM, thomas sierocinski wrote: > Dear All, > > I'm using XPS to analyse Human Exon Arrays. > I tried DABG and MAS5 to compute the detection calls and I'm getting very > different results (It seems to me that things gets scrambled at some point). > > here is the code I used: > > root.path<- "/home/thomas/QC.reporting/batch27/ROOT_files" > root.filename<- "batch2.root" > data.path<- "/home/thomas/QC.reporting/batch27/ROOT_files" > data.filename="data_batch27.root" > output.path<- "/home/thomas/QC.reporting/batch27/" > > root.data<- load.root.data(root.path, root.filename, data.path, > data.filename) > data.rma<- rma(root.data, filedir = root.path, filename = paste("tmpdt", > "RMA", root.filename, sep="_"), tmpdir = paste(getwd(), "tmp", sep="/"), > background="antigenomic", normalize = TRUE, option = "exon", exonlevel = > "all", verbose=TRUE) > call.dabg<- dabg.call(root.data, filedir = root.path, filename = > paste("tmpdt", "RMA", root.filename, sep="_"), option = "exon", exonlevel = > "all", verbose = TRUE) > call.mas5<- mas5.call(root.data, filedir = root.path, filename = > paste("tmpdt", "MAS5", root.filename, sep="_"), option = "exon", exonlevel = > "all", verbose = TRUE) > > The parameters for each detection call: >> call.dabg at params > $alpha1 > [1] 0.04 > > $alpha2 > [1] 0.06 > >> call.mas5 at params > $tau > [1] 0.015 > > $alpha1 > [1] 0.04 > > $alpha2 > [1] 0.06 > > $ignore > [1] TRUE > > The results I obtain (in matter of Present/Marginal/Absent): > PivotTable.txt using DABG: > Parameter W764.dab W766.dab > PercentAbsent 88.6888 90.155 > PercentMarginal 2.8019 2.4824 > PercentPresent 8.50933 7.36258 > > PivotTable.txt: using MAS5 > Parameter W764.dc5 W766.dc5 > PercentAbsent 9.48876 9.89098 > PercentMarginal 1.96661 1.96727 > PercentPresent 88.5446 88.1418 > > The results obtained contradicts each other. Is there any reason for that? > What am I doing wrong here? > > Thanks in advance for your answers. > > Thomas > > [[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 >