Dear BioC users, sorry to bother you, but I'm new in R/Bioconductor. I' would like to analyze microarray HG-U133A chips as from GSE3494 but using Mas5 as background correction and normalization algorithm. I use the following commands as from library(affy): abatch<- ReadAffy()#to read the cel files eset.mas5 = mas5(abatch)#to run the Mas5 algorithm exprSet.nologs = exprs(eset.mas5)#to get the expression matrix (probesets/genes in rows, chips in columns). exprSet = log(exprSet.nologs, 2)# to transform in log2 the absolute gene expression summarized values. Now the point is: after log2 transformation I found negative values even using a scaling factor of 600. What am I doing wrong? I tried to use the simpleaffy package using justMAS as function and setting a tgt = 600, but the problem still remain. Can anyone help me? thanks in advance Eleonora -- output of sessionInfo(): R version 2.12.1 (2010-12-16) Platform: x86_64-apple-darwin9.8.0/x86_64 (64-bit) -- Sent via the guest posting facility at bioconductor.org.

Hi Eleonora, On 4/3/2012 6:45 AM, Eleonora Lusito [guest] wrote: > Dear BioC users, sorry to bother you, but I'm new in R/Bioconductor. I' would like to analyze microarray HG-U133A chips as from GSE3494 but using Mas5 as background correction and normalization algorithm. I use the following commands as from library(affy): > abatch<- ReadAffy()#to read the cel files > eset.mas5 = mas5(abatch)#to run the Mas5 algorithm > exprSet.nologs = exprs(eset.mas5)#to get the expression matrix (probesets/genes in rows, chips in columns). > exprSet = log(exprSet.nologs, 2)# to transform in log2 the absolute gene expression summarized values. > > Now the point is: after log2 transformation I found negative values even using a scaling factor of 600. What am I doing wrong? I tried to use the simpleaffy package using justMAS as function and setting a tgt = 600, but the problem still remain. Can anyone help me? I think you might be misunderstanding the scale value in mas5(). The math for the scale value is new.vals = (scale.value/mean.vals) * old.vals where mean.vals are the trimmed means of the unscaled values and old.vals are the unscaled values. So you are just centering the data, and then multiplying by 600. This will not cause negative numbers to become positive. The usual prescriptive for negative mas5 numbers is to shift all data by a constant, so the smallest value is some small positive value. Something like exprs(eset) <- sweep(exprs(eset), abs(min(exprs(eset))) + 5, "+") then take logs. Best, Jim > > thanks in advance > > > Eleonora > > -- output of sessionInfo(): > > R version 2.12.1 (2010-12-16) > Platform: x86_64-apple-darwin9.8.0/x86_64 (64-bit) > > > -- > Sent via the guest posting facility at bioconductor.org. > > _______________________________________________ > 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 -- James W. MacDonald, M.S. Biostatistician University of Washington Environmental and Occupational Health Sciences 4225 Roosevelt Way NE, # 100 Seattle WA 98105-6099