[BioC] comparing Limma output from 2 different pre-processing methods

[Jenny Drnevich]

Hi Martha,

What I've done to compare the results of different analyses is to 
create a new "TestResults" object combining the output of several 
different decideTests:

 > results1 <- decideTests(fit1)
 > results2 <- decideTests(fit2)
 > results3 <- decideTests(fit3)
 > results.all <- new("TestResults",cbind(results1, results2, results3))
         # results 1-3 must have rows all in the same order

vennCounts() and vennDiagram() will now work with results.all, but 
you'll have to pick up which 3 columns if you have more than that. 
vennSelect() probably won't work (I never use it so I haven't tried), 
so if you want to create gene lists rather than number comparisons, 
you'll have to do it manually, e.g.:

 > genelist <- fit1$genes[results.all[,1] ==1 & results.all[,3] ==1, ]


Cheers,
Jenny

At 04:17 PM 11/19/2009, Martha Behnke wrote:
>Hello all,
>
>I am working with data with a very large number of differentially expressed
>genes (liver cancer samples), and wish to demonstrate for my advisor and
>committee the effect of choosing different pre-processing methods.  To do
>this, I have normalized the data in several different ways and genereated
>DEG lists with limma and decideTests.   Now I would like to examine the
>difference in which probesets are called by the various methods.
>
>I've read everything I can find about the venn tools in limma and
>vennSelect, but they only seem to compare the output of different contrasts
>within a single dataset.  Can someone suggest how to compare the output of
>the same contrast in a dataset processed 2 ways?  I could output to
>Microsoft Access, but surely there is a better way!!
>
>Thanks so much,
>Mikki Behnke
>Integrative Life Sciences
>Virginia Commonwealth University
>
>         [[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