Actually, you are close. After the lmFit function you need the eBayes, and then extract the results via topTable. See the limma manual at Bioconductor, it guides you. The stats magic is being taken care of by the function. Be sure though to do some QC on the data. In the previous post it was said that the authors did some batch correction on the data. If that is all new to you be sure to read yourself into microarray and/or RNA-seq analysis or results are unreliable and a mess. Can't you get the differential genes from the associated paper?
How can we apply Wilcoxon Mann Whitney Test for GEO data?.
This is my cording part. Can anyone help me implement the Wilcoxon Mann Whitney Test for gene data? Please help me. I really need Your help.
celFiles <- list.celfiles()
affyRaw <- read.celfiles(celFiles)
eset <- oligo ::rma(affyRaw)
write.exprs(eset,file = "data1.txt")
mydata <- read.delim("GPL570.txt",check.names = FALSE)
abc <- data.frame(mydata)
feature.data<- abc[, c(1,3)]
dataInf <- read.delim("data1.txt",check.names = FALSE)
combined <- left_join(dataInf,feature.data, by= "ID")
write.csv(combined,"anotated.csv")
pData(eset)
Group <- c("AD","AD","AD","AD","AD","AD","Normal","AD","AD","AD","AD","AD","Normal","Normal","Normal","Normal","Normal","Normal","Normal","Normal","Normal","AD","AD","AD","Normal","Normal","Normal","Normal","AD","AD","AD","AD","AD","Normal","AD","AD","AD","AD","AD","Normal","AD","AD","AD","Normal","Normal","AD")
design <- model.matrix(~factor(Group)+ 0, combined)
colnames(design) <- c("AD", "Normal")
fit <- lmFit(eset, design)
• 2,060 views
•
link
1 answer
Wilcox tests are not appropriate for this as already pointed out here (Hi everyone. I want to know how to analysis genes data by using "Welch t test, Wilcoxon Mann Whitney test(WMW) and Significance of Analysis Microarray(SAM) " With r codes.). Just use limma. It has a comprehensive manual. Don't do custom approaches without experts knowledge, use limma.
• 0 views
•
link
• 0 views
•
link
Log in to answer this question.