I'm fairly new to Bioconductor and had a basic question about *recommended* steps to analyze replicate data. I apologize if this has been covered, but I've scoured the message archives and couldn't find any examples to guide me. I'd like to look at 3 biological replicates of treated vs. untreated cells and determine p-values for over- or under-expression of each gene. We're using in-house spotted long oligo two-color microarrays and scanning with GenePix. I've read in my experiment layout, gal file, and gpr data using: ---------------------------------------------------------------------- -- - exp.layout<- read.marrayLayout(fname=galfile,ngr=8,ngc=4,nsr=21,nsc=20,skip=37,ctl. co l=6); exp.targets<-read.marrayInfo(fname=layoutfile) exp.genes<-read.marrayInfo(fname=galfile,labels=4,skip=37) exp.raw<-read.marrayRaw(fnames,path=datadir,exp.Gf="F532 Median",exp.Gb="B532 Median",exp.Rf="F635 Median",exp.Rb="B635 Median",exp.W="Flags",skip=30,layout=exp.layout,gnames=exp.genes,targe ts =exp.targets) : : other filtering here : : exp.norm <- maNorm(exp.filtered,n="s") ---------------------------------------------------------------------- -- - I've also generated LOWESS normalized data by spotting pin and now would like to determine the significance of changes I see. I've seen limma and multtest for doing this but have been unable to determine if these functions are compatible with how I've imported my data or how to call them correctly. Can anyone point me in the right direction given that I have my filtered/normalized data from the three runs in a variable called exp.norm? Also, should I normalize ACROSS arrays in addition to within arrays prior to the p-value calculation? What syntax is used for this? Thanks. -- Mike Schaffer