[BioC] help in creating custom venn diagrams from multiple contrasts

[James W. MacDonald]

Hi Triantafyllos,

Triantafyllos Paparountas wrote:
> Hi all,
> 
> I am trying to run an analysis on 7 groups of chips (1 ctrl and 6
> different conditions), each with 2 affymetrix chips.
> 
> I am trying to replicate what affylm gui would do but instead I
> thought of using a gcrma.mle correction.
> 
> After gathering as much info as I could from the forum ,still it
> seems to have stuck on how to choose which contrasts to use for the
> venn diagrams. (see part "#create vennDiagram" below).
> 
> What I could consider important here is the ability to choose the
> contrasts that would be shown on the venn diagram , in a way that
> affylmGUI would enable the user to use.
> 
> Any help would really be appreciated.
> 
> ---------- By the code used below I try to make a huge vennDiagram of
> all the 21 contrasts which is why possibly R returns"cannot allocate
> a vector of 340000 kb". Anyhow this would be a nice thing to have ,
> possibly not usefull for the work I need to do ,  but just for the
> sake of seeing if this is possible to be done. ----------

limma is not capable of creating Venn diagrams of greater than three 
comparisons. Although it is possible to do higher order Venn diagrams, I 
don't think they would be easily programmed. See here for more information:

http://en.wikipedia.org/wiki/Venn_diagram

> 
> 
> # start by loading base libraries library(affy) library(limma) 
> library(gcrma)
> 
> #read in files
> 
> targets <- readTargets("RT_target.txt", row.names="Name") 
> ex1<-ReadAffy(filenames=targets$FileName) 
> ex1.gcrma.mle<-gcrma(ex1,estimate="mle",normalize=TRUE,summary.method
> = "medianpolish" )
> 
> #create lin model and contrasts design <- model.matrix(~
> -1+factor(c(1,1,2,2,3,3,4,4,5,5,6,6,7,7))) colnames(design) <-
> c("CTRL", "IGF1", "ZOME","TGFb","CTED","INT6","ILTI") fit <-
> lmFit(ex1.gcrma.mle, design) contrast.matrix <-
> makeContrasts(IGF1-CTRL, ZOME-CTRL,
> TGFb-CTRL,CTED-CTRL,INT6-CTRL,ILTI-CTRL,ZOME - IGF1,
> TGFb-IGF1,CTED-IGF1,INT6-IGF1,ILTI-IGF1,TGFb-ZOME,CTED-ZOME,INT6-ZOME,ILTI-ZOME,CTED-TGFb,INT6-TGFb,ILTI-TGFb,INT6-CTED,ILTI-CTED,ILTI-INT6,
> levels=design) fit2 <- contrasts.fit(fit, contrast.matrix) fit2 <-
> eBayes(fit2) topTable(fit2, coef=1, adjust="BH")
> 
> #create vennDiagram results <- decideTests(fit2) results 
> vennDiagram(results)

This should just error out, saying that you cannot make a Venn diagram 
with greater than three contrasts.

However, you should be able to make a Venn diagram with up to three 
contrasts by simply subsetting your 'results' matrix. For instance

vennDiagram(results[,1:3])

HTH,

Jim


> 
> 
> 
> #Export toptable library(affycoretools) 
> limma2annaffy(ex1,fit2,design,contrast.matrix, lib="hgu133a",adjust =
> "fdr",anncols = aaf.handler()[c(1:13)], number = 30, pfilt = NULL,
> fldfilt= NULL,tstat = TRUE, pval = TRUE, FC = TRUE,expression = TRUE,
> html = TRUE, text = FALSE, save = FALSE, addname =NULL, interactive =
> TRUE)
> 
> 
> [[alternative HTML version deleted]]
> 
> _______________________________________________ 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


-- 
James W. MacDonald
University of Michigan
Affymetrix and cDNA Microarray Core
1500 E Medical Center Drive
Ann Arbor MI 48109
734-647-5623



**********************************************************
Electronic Mail is not secure, may not be read every day, and should not be used for urgent or sensitive issues.