[BioC] venn diagram

[Jenny Drnevich]

Hi Stephen,

Sean's suggestion of summary(result) will give you the individual sig 
up and sig down counts for each contrast. I looked at the help for 
"TestResults-class" and what it doesn't explicitly say either is that 
this object is basically a matrix of 1s, 0s or -1s. You can see this 
by looking at the first few rows:

 > results[1:10,]

To use it to find the probe sets unique in each condition, you can do:

 >  unique1 <- sampleNames(x) [results[,1] !=0 & results[,2]==0 & 
results[,3]==0]
 >  unique2 <- sampleNames(x) [results[,1] ==0 & results[,2]!=0 & 
results[,3]==0]
 >  unique3 <- sampleNames(x) [results[,1] ==0 & results[,2]==0 & 
results[,3]!=0]

HTH,
Jenny

At 07:40 AM 7/30/2009, Sean Davis wrote:
>On Thu, Jul 30, 2009 at 2:13 AM, Stephen V. Su <svs8 at yahoo.com> wrote:
>
> > To whom this may concern:
> >
> >     I must first apologize for not being a bioinformaticist and really only
> > "pretend" to know how to use R to analyze microarray.  But i do understand
> > the concepts of Limma and when to apply it.  I think it is a really
> > powerful way to analyze matrix data.  So, if you don't mind, I 
> would like to
> > ask you a question regarding Venn Diagrams.  I have generated a
> > RMA-processed expression set for 3 different conditions (each condition has
> > a minimum of 8 CEL files- Affy 0.5 genome rat chip), fed it 
> through limma to
> > create a statistical computation of all the genes that are differentially
> > expressed and created a venn Diagram to depict the results.  A 
> general input
> > functions of what i just wrote is depicted here:
> >
> >
> > > fit <- lmFit(x, design)
> >
> > > cont.matrix <- makeContrasts(x3Dvsinvivo = x3D - invivo,
> > + x3Dvsx2D = x3D - x2D,
> > + x2Dvsinvivo = x2D - invivo, levels = design)
> > > fit2 <- contrasts.fit(fit, cont.matrix)
> > > fit2 <- eBayes(fit2)
> > > results <- decideTests(fit2)
> > > vennDiagram(results)
> >
> > My question is:  How do I extract out the genes that reside in each of the
> > quadrant in the Venn diagram?  Is there a way for me to say, retrieve the
> > 108 genes that are unique to condition A, 98 genes unique to 
> condition B and
> > C, and so on ....
>
>
>Hi, Steve.  The answer is in the help system.  If you read the help for
>decideTests, you will see that the value from a call to decideTests is an
>object of class "TestResults".  Doing help("TestResults-class") will tell
>you that an object of class "TestResults" can be treated like a matrix.  It
>also has a few methods, one of which is the summary method.  So, try:
>
>summary(result)
>
>Does that help?
>
>Sean
>
>         [[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

Jenny Drnevich, Ph.D.

Functional Genomics Bioinformatics Specialist
W.M. Keck Center for Comparative and Functional Genomics
Roy J. Carver Biotechnology Center
University of Illinois, Urbana-Champaign

330 ERML
1201 W. Gregory Dr.
Urbana, IL 61801
USA

ph: 217-244-7355
fax: 217-265-5066
e-mail: drnevich at illinois.edu