[BioC] question about the code in Lab 4 - Differential Expression and Linear Modeling using limma(Estrogen Data)

[Gordon Smyth]

>[BioC] question about the code in Lab 4 - Differential Expression and 
>Linear Modeling using limma(Estrogen Data) on 
>http://www.bioconductor.org/workshops/2005/BioC2005/labs/lab01/estrogen.html
>Na, Ren Na at uthscsa.edu
>Tue Nov 15 00:21:10 CET 2005
>
>Hi,
>I tried the code in section 11 "Gene Set Enrichment" on 
>http://www.bioconductor.org/workshops/2005/BioC2005/labs/lab01/estrogen.html, 
>the original code is:
>1  known <- read.delim("knownERgenes.txt",as.is=TRUE)
>2  knownERgenes <- known$UGCluster
>3  predicted <- read.delim("predictedERgenes.txt",as.is=TRUE)
>4  predictedERgenes <- predicted$UGCluster
>5  library(hgu95av2)
>6  unigene <- unlist(as.list(hgu95av2UNIGENE))
>7  knownERgenesOnChip <- match(knownERgenes,unigene)
>8  knownERgenesOnChip <- knownERgenesOnChip[!is.na(knownERgenesOnChip)]
>9  predictedERgenesOnChip <- match(predictedERgenes,unigene)
>10 predictedERgenesOnChip <- 
>predictedERgenesOnChip[!is.na(predictedERgenesOnChip)]
>11 geneSetTest(knownERgenesOnChip,fit2$t[,1],"two.sided")
>
>Please correct me if I am wrong.
>I think line 6 should be replaced by the following two lines:
>geneIDs <- ls(hgu95av2cdf)
>unigene <- as.character(unlist(lapply(mget(geneIDs,env=hgu95av2UNIGENE),
>      function (unigeneID) { return(paste(unigeneID,collapse="; ")) } )))
>the order of "unigene" from line 6 is not same as the order of the above 
>"unigene"
>
>otherwise knownERgenesOnChip is not indices of knownERgenes on fit2$t[,1].
>
>Thanks in advance!
>Ren

I guess you are right. I've put a corrected version of the Lab at 
http://bioinf.wehi.edu.au/marray/bioc2005/estrogen.html

Best wishes
Gordon