Hello bioconductors, while trying to make a vennDiagram I got the following error: > results <- decideTests(eb, method="separate",adjust.method="fdr",p.value=0.05) > results TestResults matrix PPK PRO PCD [1,] 0 0 -1 [2,] -1 1 0 [3,] 0 1 0 [4,] -1 0 -1 [5,] -1 0 0 6139 more rows ... > vennDiagram(results,method="separate",include="both") Error in toptable(fit = fit[c("coefficients", "stdev.unscaled")], coef = coef, : (list) object cannot be coerced to integer Has anyone seen this before or know how to fix it? I think I passed the right arguments (?) based on the help files. I'm using R 2.0.1 and Bioconductor release 1.5. Thanks -Dennis

Hi Dennis, Since vennDiagram is in limma, it would help to add the version you are using to your post. The most recent version of limma is available on CRAN (not Bioconductor) and it might be worth upgrading and makeing sure this hasn't already been fixed. + seth

I was using limma 1.8.16, but I took your advice and updated to 1.8.22 from R-CRAN. That did not solve the problem (see previous post) unfortunately. Is there anything suspicious about my TestResults object or is there more info I can provide? This is perplexing to me. I've tried running simpler model structures (i.e., one coefficient instead of three) and various different calls of decideTests() and vennDiagram() and I still get this message. I recently installed both R and bioconductor from scratch and have not monkeyed with anything. I use Debian linux 2.6.4 if that helps.

> Date: Tue, 22 Mar 2005 22:57:27 -0800 > From: Dennis Hazelett <hazelett@uoneuro.uoregon.edu> > Subject: [BioC] vennDiagram error: Error in toptable > To: bioconductor@stat.math.ethz.ch > > Hello bioconductors, > while trying to make a vennDiagram I got the following error: > > > results <- decideTests(eb, > method="separate",adjust.method="fdr",p.value=0.05) > > results > TestResults matrix > PPK PRO PCD > [1,] 0 0 -1 > [2,] -1 1 0 > [3,] 0 1 0 > [4,] -1 0 -1 > [5,] -1 0 0 > 6139 more rows ... > > vennDiagram(results,method="separate",include="both") > Error in toptable(fit = fit[c("coefficients", "stdev.unscaled")], coef = > coef, : > (list) object cannot be coerced to integer Your email seems to be garbled. Firstly, vennDiagram() doesn't have an argument 'method'. Secondly, the error message seems to be from quite a different function call -- there is no way that a call to vennDiagram() can lead to a call to toptable and to the error message given. The call given in the error message itself looks quite strange, and is not something which appears anywhere in the limma package. You certainly don't seem to be using the commands correctly, or reporting correctly what has happened. Gordon > Has anyone seen this before or know how to fix it? I think I passed the > right arguments (?) based on the help files. I'm using R 2.0.1 and > Bioconductor release 1.5. > Thanks > -Dennis