[BioC] Inf values using SAM (package=siggenes)
Holger Schwender
holger.schw at gmx.de
Sun Jan 6 17:53:39 CET 2008
Dear Julian,
the only values that are infinitive are the values of cutlow and cutup, where cutup is the smallest value of the test statistic for a gene to be called differentially expressed if its test value is larger than zero, and cutup is the largest test score less than zero for a gene to be called differentially expressed. If none of the genes is called differentially expressed by SAM than cutlow and cutup are set to -Inf and Inf, respectively. This has both practical and theoretical reasons.
Best,
Holger
-------- Original-Nachricht --------
> Datum: Fri, 28 Dec 2007 11:43:33 +0800 (SGT)
> Von: Julian Lee <julian at omniarray.com>
> An: bioconductor at stat.math.ethz.ch
> Betreff: [BioC] Inf values using SAM (package=siggenes)
> Dear All,
>
> I'm having some problems trying to find differentially expressed genes
> using SAM. I have the SAM for excel version but that unfortunately is limited
> to only 500 genes.
>
> >from the vignette, i'm supplying the sam function, two important
> arguments, data and cl.
>
> > data
> ExpressionSet (storageMode: lockedEnvironment)
> assayData: 16304 features, 65 samples
> element names: exprs
> phenoData
> rowNames: D02_2nd, D02_3rd, ..., D31_BL (65 total)
> varLabels and varMetadata:
> Patient_ID: Patient's ID
> Patient_Initials: Patient's Initials
> ...: ...
> 25_at_cycle1: 25% reduction from baseline at cycle1
> (17 total)
> featureData
> rowNames: 1007_s_at, 1053_at, ..., AFFX-r2-Ec-bioD-5_at (16304 total)
> varLabels and varMetadata: none
> experimentData: use 'experimentData(object)'
> Annotation [1] "hgu133plus2"
>
> >cl
> [1] 1 1 0 1 1 0 1 1 0 1 0 1 1 1 1 1 0 1 1 0 1 1 0 1 1 0 1 1 0 1 1 1 0 1 1
> 0 1 1
> [39] 1 1 0 1 1 0 1 1 0 1 1 1 0 1 0 1 1 0 1 1 0 1 0 1 0 1 0
>
> data was pre-processed using the rma function, followed by genefiltering
>
> >data<-rma(U133PLUS2 CELFILES)
> >library(genefilter)
> >f1<-pOverA(0.25,log2(100))
> >f2<-function(x) (IQR(x)>0.5)
> >ff<-filterfun(f1,f2)
> >data<-data[genefilter(data,ff),]
>
> ##56,000 genes reduced to 16304 genes
>
> >library(siggenes)
>
> >sam.out<-sam(exprs(data),cl,rand=1234)
> >sam.out
> SAM Analysis for the Two-Class Unpaired Case Assuming Unequal Variances
>
> Delta p0 False Called FDR
> 1 0.1 1 0 0 0
> 2 0.2 1 0 0 0
> 3 0.3 1 0 0 0
> 4 0.4 1 0 0 0
> 5 0.5 1 0 0 0
> 6 0.6 1 0 0 0
> 7 0.7 1 0 0 0
> 8 0.8 1 0 0 0
> 9 0.9 1 0 0 0
> 10 1.0 1 0 0 0
>
> >summary(sam.out)
> SAM Analysis for the Two-Class Unpaired Case Assuming Unequal Variances
>
> s0 = 0.0735 (The 0 % quantile of the s values.)
>
> Number of permutations: 100
>
> MEAN number of falsely called variables is computed.
>
> Delta p0 False Called FDR cutlow cutup j2 j1
> 1 0.1 1 0 0 0 -Inf Inf 0 16305
> 2 0.2 1 0 0 0 -Inf Inf 0 16305
> 3 0.3 1 0 0 0 -Inf Inf 0 16305
> 4 0.4 1 0 0 0 -Inf Inf 0 16305
> 5 0.5 1 0 0 0 -Inf Inf 0 16305
> 6 0.6 1 0 0 0 -Inf Inf 0 16305
> 7 0.7 1 0 0 0 -Inf Inf 0 16305
> 8 0.8 1 0 0 0 -Inf Inf 0 16305
> 9 0.9 1 0 0 0 -Inf Inf 0 16305
> 10 1.0 1 0 0 0 -Inf Inf 0 16305
>
> I'm not too sure why i'm getting Infinity values. This is my first time
> using SAM on bioconductor.
>
> Thank you
>
> regards
>
> Julian Lee
> National Cancer Center Singapore
>
> >sessionInfo()
> R version 2.5.1 (2007-06-27)
> i386-pc-mingw32
>
> locale:
> LC_COLLATE=English_United States.1252;LC_CTYPE=English_United
> States.1252;LC_MONETARY=English_United States.1252;LC_NUMERIC=C;LC_TIME=English_United
> States.1252
>
> attached base packages:
> [1] "splines" "tools" "stats" "graphics" "grDevices" "utils"
> [7] "datasets" "methods" "base"
>
> other attached packages:
> genefilter maDB limma affy affyio siggenes
> multtest
> "1.14.1" "1.8.0" "2.10.5" "1.14.2" "1.4.1" "1.10.1"
> "1.16.1"
> survival Biobase
> "2.32" "1.14.1"
>
> _______________________________________________
> 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
--
More information about the Bioconductor
mailing list