Analyzing data from Agilent-023795 Human DNA Methylation Microarray 244k

Question

[emiliomastriani]

Dear all, It's the first time that I move to analyze methylated data. I will start with the dataset https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE96067, containing data extracted from Agilent-023795 Human DNA Methylation Microarray 244k (Probe Name Version). My goal is to look for differentially methylated regions and related genes. Please, can someone suggest me which package will be suitable for my scope? Any suggest is welcome. Thank you guys. Emilio

Answer 1

[Kevin Blighe]

On the page to which you link, click on the blue button called Analyze with GEO2R. Then, click on the R script tab and execute the code that appears. That will, more than likely, give you the normalised beta methylation values (check via boxplot). With those, you can do the differential methylation analysis.

That is:

library(Biobase)
library(GEOquery)

# load series and platform data from GEO
gset <- getGEO("GSE96067", GSEMatrix =TRUE, getGPL=FALSE)
if (length(gset) > 1) idx <- grep("GPL19930", attr(gset, "names")) else idx <- 1
gset <- gset[[idx]]

# set parameters and draw the plot
dev.new(width=4+dim(gset)[[2]]/5, height=6)
par(mar=c(2+round(max(nchar(sampleNames(gset)))/2),4,2,1))
title <- paste ("GSE96067", '/', annotation(gset), " selected samples", sep ='')
boxplot(exprs(gset), boxwex=0.7, notch=T, main=title, outline=FALSE, las=2)

Kevin

[emiliomastriani]

Thanks Kevin. Just another question: which one is the better package to make the differential methylation analysis in may case?

[Kevin Blighe]

You can likely use the Wilcoxon Signed Rank test if you are comparing samples that are 'generally' similar/related. or those that are paired - it is a non-parametric test, which is the safest bet in these cases. Typically, in methylation, we obtain a p-value from a test and also the difference in mean methylation.

So, Difference in means = mean methylation Group A - mean methylation Group B