[BioC] Rgraphviz, R 2.7.0 and Windows XP - SOLVED!

Jenny Drnevich drnevich at illinois.edu
Wed Jun 11 23:30:34 CEST 2008


At 03:52 PM 6/11/2008, James W. MacDonald wrote:
>Martin just helped me figure out where I am missing things. I _do_ 
>have Graphviz_2.16.1 'installed' in my program files directory. I 
>say 'installed' because I evidently just stuck it in there and put 
>it in my path.
>
>I wonder if your problem has to do with the version of Graphviz then?

Could be... Early on in this long thread, it was said that Graphviz 
2.16.1 specifically was needed, because that's what was used to make 
the Windows binary of Rgraphviz. However, when I tried it on the 
first virgin PC yesterday, I was still getting a Windows error saying 
"The application has failed to start because somefilename.dll was not 
found..."  Then I uninstalled Graphviz 2.16.1 and installed Graphviz 
2.18.1 and that seemed to work. Then I did everything, including 
installing R 2.7.0 from scratch, on another PC and only ever 
installed Graphviz 2.18.1 and was able to load the Rgraphviz library 
without any Windows or R errors.  But I guess it's still not working 
correctly. SIGH.

What now?
Jenny



>Jim
>
>
>
>James W. MacDonald wrote:
>>Yeah I remember seeing that and assuming I had already installed 
>>Graphviz at some point in time. However, upon inspection of my comp 
>>I don't see any evidence that I have done so.
>>I usually build R from source because I think zlib works better 
>>that way (or some such reason that I have forgotten by now), and I 
>>then put it right in C:\ (e.g., C:\R-2.7.0 or whatever). I wonder 
>>if the path matters?
>>Jim
>>
>>Jenny Drnevich wrote:
>>>At 02:39 PM 6/11/2008, James W. MacDonald wrote:
>>>>I just updated to Rgraphviz_1.18.1 (the binary version) and it 
>>>>still works for me. Plus, I have never installed Graphviz on this 
>>>>box, nor the  Visual Studio redistributable.
>>>
>>>REALLY!?!  Some of us were having problems getting the binary to 
>>>work back in April after switching to R 2.7.0. Here's Martin's 
>>>original post saying it was because we needed to install Graphviz:
>>>
>>>https://stat.ethz.ch/pipermail/bioconductor/2008-April/022026.html
>>>
>>>I'll have to see if I can find another virgin PC to test... good 
>>>thing I work in a core lab :)
>>>
>>>Jenny
>>>
>>>
>>>
>>>>I haven't followed this thread closely since 
>>>>biocLite("Rgraphviz") has always worked for me out of the box, 
>>>>but is it really necessary to have these extras installed for the 
>>>>binary version?
>>>>
>>>>Jim
>>>>
>>>>
>>>>
>>>>Jenny Drnevich wrote:
>>>>>Hi Jim,
>>>>>Rgraphviz_1.18.1 . Did you build Rgraphviz from source? If so, 
>>>>>then the problem may be related to using the Windows binary, not the OS.
>>>>>Jenny
>>>>>At 12:14 PM 6/11/2008, James W. MacDonald wrote:
>>>>>>What version of Rgraphviz is this? I don't have any problem 
>>>>>>running Sweave on the topGO vignette:
>>>>>>
>>>>>> > Sweave("topGO.Rnw")
>>>>>>Writing to file topGO.tex
>>>>>>Processing code chunks ...
>>>>>>  1 : term verbatim
>>>>>>  2 : echo term hide
>>>>>>  3 : echo term verbatim
>>>>>>  4 : echo term verbatim
>>>>>>  5 : echo term verbatim
>>>>>>Loading required package: survival
>>>>>>Loading required package: splines
>>>>>>  6 : echo term hide
>>>>>>  7 : echo term verbatim
>>>>>>  8 : echo term verbatim
>>>>>>  9 : echo term verbatim
>>>>>>10 : echo term hide
>>>>>>11 : echo term hide
>>>>>>Loading required package: multtest
>>>>>>12 : term verbatim eps pdf
>>>>>>13 : echo term hide
>>>>>>14 : echo term verbatim
>>>>>>15 : echo term hide
>>>>>>16 : echo term hide
>>>>>>17 : echo term hide
>>>>>>18 : echo term hide
>>>>>>19 : echo term hide
>>>>>>20 : echo term hide
>>>>>>21 : echo term verbatim
>>>>>>22 : echo term hide
>>>>>>23 : echo term hide
>>>>>>24 : echo term verbatim
>>>>>>25 : echo term verbatim
>>>>>>26 : echo term verbatim
>>>>>>27 : echo term verbatim
>>>>>>28 : echo term verbatim
>>>>>>29 : echo term verbatim
>>>>>>30 : term verbatim
>>>>>>31 : term tex
>>>>>>Loading required package: xtable
>>>>>>32 : echo term hide
>>>>>>Loading required package: Rgraphviz
>>>>>>Loading required package: grid
>>>>>>33 : term hide
>>>>>>34 : echo term verbatim
>>>>>>35 : echo term verbatim
>>>>>>36 : term tex
>>>>>>
>>>>>>You can now run LaTeX on 'topGO.tex'
>>>>>> > sessionInfo()
>>>>>>R version 2.7.0 (2008-04-22)
>>>>>>i386-pc-mingw32
>>>>>>
>>>>>>locale:
>>>>>>LC_COLLATE=English_United States.1252;LC_CTYPE=English_United 
>>>>>>States.1252;LC_MONETARY=English_United 
>>>>>>States.1252;LC_NUMERIC=C;LC_TIME=English_United States.1252
>>>>>>
>>>>>>attached base packages:
>>>>>>  [1] grid      splines   tools     stats     graphics  grDevices
>>>>>>  [7] datasets  utils     methods   base
>>>>>>
>>>>>>other attached packages:
>>>>>>  [1] Rgraphviz_1.18.0    xtable_1.5-2        multtest_1.20.0
>>>>>>  [4] genefilter_1.20.0   survival_2.34-1     hgu95av2.db_2.2.0
>>>>>>  [7] ALL_1.4.3           topGO_1.8.1         SparseM_0.77
>>>>>>[10] GO.db_2.2.0         AnnotationDbi_1.2.0 RSQLite_0.6-8
>>>>>>[13] DBI_0.2-4           Biobase_2.0.1       graph_1.18.1
>>>>>>
>>>>>>loaded via a namespace (and not attached):
>>>>>>[1] annotate_1.18.0 cluster_1.11.10
>>>>>> >
>>>>>>
>>>>>>Jenny Drnevich wrote:
>>>>>>>HI Sebastien,
>>>>>>>It would be more helpful if you let us know at what point 
>>>>>>>during the vignette code your Rgui crashed. I also was not 
>>>>>>>able to get through the vignette - I got up through page 9, 
>>>>>>>and the crash occurred when I called
>>>>>>>  > showSigOfNodes(GOdata, score(resultFis), firstTerms = 5,
>>>>>>>+ useInfo = "all")
>>>>>>>Stepping through the showSigOfNodes function, the problem 
>>>>>>>occurs with the (internal?) function plotFunction():
>>>>>>>  >complete.dag <- plotFunction(dag, sigNodes = sigNodes, 
>>>>>>> genNodes = names(sigTerms),
>>>>>>>         wantedNodes = wantedNodes, showEdges = showEdges, 
>>>>>>> useFullNames = useFullNames,
>>>>>>>         oldSigNodes = oldSigNodes, nodeInfo = nodeInfo)
>>>>>>>It's not easy to figure out what plotFunction is doing, and 
>>>>>>>that's all the time I have for it now. It may be the same 
>>>>>>>known problem that causes
>>>>>>>  > example(agwrite)
>>>>>>>to also crash R. For the life of me, I can't remember exactly 
>>>>>>>what the problem is, and don't have time to search through the 
>>>>>>>archives to find the post by the wonderful person who pointed it out!
>>>>>>>Below is my sessionInfo(), right before showSigOfNodes() is 
>>>>>>>called. showSigOfNodes() does require Rgraphviz...
>>>>>>>Cheers,
>>>>>>>Jenny
>>>>>>>  > allRes <- GenTable(GOdata, classic = resultFis, KS = resultKS,
>>>>>>>+ elim = resultElim, weight = resultWeight, orderBy = "weight",
>>>>>>>+ ranksOf = "classic", topNodes = 20)
>>>>>>>  > sessionInfo()
>>>>>>>R version 2.7.0 (2008-04-22)
>>>>>>>i386-pc-mingw32
>>>>>>>locale:
>>>>>>>LC_COLLATE=English_United States.1252;LC_CTYPE=English_United 
>>>>>>>States.1252;LC_MONETARY=English_United 
>>>>>>>States.1252;LC_NUMERIC=C;LC_TIME=English_United States.1252
>>>>>>>attached base packages:
>>>>>>>[1] splines   tools     stats     graphics  grDevices utils
>>>>>>>datasets
>>>>>>>[8] methods   base
>>>>>>>other attached packages:
>>>>>>>  [1] multtest_1.20.0     genefilter_1.20.0   survival_2.34-1
>>>>>>>  [4] hgu95av2.db_2.2.0   ALL_1.4.3           topGO_1.8.1
>>>>>>>  [7] SparseM_0.78        GO.db_2.2.0         AnnotationDbi_1.2.0
>>>>>>>[10] RSQLite_0.6-8       DBI_0.2-4           Biobase_2.0.1
>>>>>>>[13] graph_1.18.1
>>>>>>>loaded via a namespace (and not attached):
>>>>>>>[1] annotate_1.18.0 cluster_1.11.10
>>>>>>>  >
>>>>>>>At 08:56 PM 6/10/2008, Sebastien Gerega wrote:
>>>>>>>>Hi,
>>>>>>>>I followed these instructions and am now able to load Rgraphviz without
>>>>>>>>error. However, when I attempt to use it R crashes.
>>>>>>>>I am not directly using Rgraphviz but am using the topGO 
>>>>>>>>package which is
>>>>>>>>calling it. I am basically just running through the example 
>>>>>>>>in the topGO
>>>>>>>>vignette when R crashes.
>>>>>>>>
>>>>>>>>Here is my sessionInfo():
>>>>>>>>R version 2.7.0 (2008-04-22)
>>>>>>>>i386-pc-mingw32
>>>>>>>>
>>>>>>>>locale:
>>>>>>>>LC_COLLATE=English_Australia.1252;LC_CTYPE=English_Australia.1252;LC_MONETARY=English_Australia.1252;LC_NUMERIC=C;LC_TIME=English_Australia.1252 
>>>>>>>>
>>>>>>>>
>>>>>>>>attached base packages:
>>>>>>>>[1] grid      tools     stats     graphics  grDevices utils
>>>>>>>>datasets
>>>>>>>>methods   base
>>>>>>>>
>>>>>>>>other attached packages:
>>>>>>>>  [1] Rgraphviz_1.18.1    topGO_1.8.1         SparseM_0.77
>>>>>>>>GO.db_2.2.0         AnnotationDbi_1.2.0 RSQLite_0.6-8
>>>>>>>>DBI_0.2-4
>>>>>>>>  [8] graph_1.18.0        arrayQuality_1.18.0 RColorBrewer_1.0-2
>>>>>>>>gridBase_0.4-3      hexbin_1.14.0       lattice_0.17-6
>>>>>>>>convert_1.16.0
>>>>>>>>[15] Biobase_2.0.0       marray_1.18.0       limma_2.14.4
>>>>>>>>GDD_0.1-12
>>>>>>>>
>>>>>>>>loaded via a namespace (and not attached):
>>>>>>>>[1] cluster_1.11.10
>>>>>>>>
>>>>>>>>thanks,
>>>>>>>>Sebastien
>>>>>>>>
>>>>>>>>
>>>>>>>>
>>>>>>>>On Wed, Jun 11, 2008 at 3:03 AM, Jenny Drnevich <drnevich at illinois.edu>
>>>>>>>>wrote:
>>>>>>>>
>>>>>>>> > Hi everyone,
>>>>>>>> >
>>>>>>>> > I finally got Rgraphviz to work from the Windows binary. 
>>>>>>>> For some reason,
>>>>>>>> > Graphviz versions that didn't seem to work last week are 
>>>>>>>> working this week!
>>>>>>>> > I just tried the following on two "virgin" Windows 
>>>>>>>> machines, and it seems to
>>>>>>>> > work:
>>>>>>>> >
>>>>>>>> > 1. Download and install Microsoft Visual C++ 2005 SP1 
>>>>>>>> Redistributable
>>>>>>>> > Package:
>>>>>>>> > 
>>>>>>>> http://www.microsoft.com/downloads/details.aspx?familyid=200B2FD9-AE1A-4A14-984D-389C36F85647&displaylang=en 
>>>>>>>>
>>>>>>>> >
>>>>>>>> > 2. Download and install the current stable release for 
>>>>>>>> Windows of Graphviz
>>>>>>>> > 2.18.1: 
>>>>>>>> http://www.graphviz.org/pub/graphviz/ARCHIVE/graphviz-2.18.1. 
>>>>>>>> exe (y es , I know before we were saying 2.16.1 was needed, but for some reason
>>>>>>>> > 2.18.1 is working this week!)
>>>>>>>> >
>>>>>>>> >
>>>>>>>> > 3. From within R, download and install Rgraphviz using:
>>>>>>>> > > source("http://bioconductor.org/biocLite.R")
>>>>>>>> > > biocLite("Rgraphviz")
>>>>>>>> >
>>>>>>>> > 4. If everything goes well, this should work:
>>>>>>>> > > library(Rgraphviz)
>>>>>>>> >
>>>>>>>> > I'm not having any troubles now, at least in loading the package!
>>>>>>>>Special
>>>>>>>> > thanks to Martin Morgan and Tony Chiang for helping me sort this out
>>>>>>>> > off-list.
>>>>>>>> >
>>>>>>>> > Cheers,
>>>>>>>> > Jenny
>>>>>>>> >
>>>>>>>> > BTW - there's a known bug in graphviz-2.18.exe that the 
>>>>>>>> uninstall.exe file
>>>>>>>> > doesn't work. Won't matter until you try to uninstall it! 
>>>>>>>> They said to
>>>>>>>> > install the development release to the same folder and 
>>>>>>>> start menu and then
>>>>>>>> > uninstall.
>>>>>>>> >
>>>>>>>> >
>>>>>>>> > At 03:51 AM 6/4/2008, Li Long wrote:
>>>>>>>> >
>>>>>>>> >> >
>>>>>>>> >> >    In the short - and likely even medium term that may 
>>>>>>>> be the only
>>>>>>>> >> > solution.  There do not seem to be easy ways to build a 
>>>>>>>> DLL that is
>>>>>>>> >> > compliant (which is why the very old one was retained 
>>>>>>>> for such a long
>>>>>>>> >> > time) and it requires a substantial effort with every 
>>>>>>>> new version of
>>>>>>>> >> > Graphviz that comes out - given competing needs for 
>>>>>>>> scarce resources, I
>>>>>>>> >> > am afraid that this is likely to be a month or more.
>>>>>>>> >> >
>>>>>>>> >> >    best wishes
>>>>>>>> >> >     Robert
>>>>>>>> >> >
>>>>>>>> >>
>>>>>>>> >> Hi,
>>>>>>>> >>
>>>>>>>> >> Graphviz releases are also not as stable as one would hope:
>>>>>>>> >>
>>>>>>>> >> (1) graphviz 2.16 didn't contain some required .dlls as 
>>>>>>>> graphviz 2.15,
>>>>>>>> >> which causes a lot of confusions, it seems like graphviz 
>>>>>>>> 2.18 contains all
>>>>>>>> >> the .dlls needed and it works ok in linking;  the README 
>>>>>>>> was written when
>>>>>>>> >> testing was done for graphviz 2.15, assuming graphviz releases are
>>>>>>>> >> somewhat stable...
>>>>>>>> >>
>>>>>>>> >> (2) graphviz built on Windows uses Microsoft Visual 
>>>>>>>> Studio, not minGW or
>>>>>>>> >> any thing Linux-like, and the developers strongly discourage
>>>>>>>> >> build-your-own approach; when mixing the .obj from these different
>>>>>>>> >> sources, things are pretty nasty, and we (me and graphviz 
>>>>>>>> developers)
>>>>>>>> >> haven't found a way out; it seems that the suggestion 
>>>>>>>> from graphviz users
>>>>>>>> >> and developers is to use .dlls;
>>>>>>>> >>
>>>>>>>> >> It's not without trying...
>>>>>>>> >>
>>>>>>>> >> Li
>>>>>>>> >>
>>>>>>>> >> _______________________________________________
>>>>>>>> >> 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
>>>>>>>> >
>>>>>>>> > _______________________________________________
>>>>>>>> > 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
>>>>>>>> >
>>>>>>>>
>>>>>>>>         [[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
>>>>>>>_______________________________________________
>>>>>>>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, M.S.
>>>>>>Biostatistician
>>>>>>Affymetrix and cDNA Microarray Core
>>>>>>University of Michigan Cancer Center
>>>>>>1500 E. Medical Center Drive
>>>>>>7410 CCGC
>>>>>>Ann Arbor MI 48109
>>>>>>734-647-5623
>>>>>_______________________________________________
>>>>>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, M.S.
>>>>Biostatistician
>>>>Affymetrix and cDNA Microarray Core
>>>>University of Michigan Cancer Center
>>>>1500 E. Medical Center Drive
>>>>7410 CCGC
>>>>Ann Arbor MI 48109
>>>>734-647-5623
>>>>
>>>>_______________________________________________
>>>>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
>
>--
>James W. MacDonald, M.S.
>Biostatistician
>Affymetrix and cDNA Microarray Core
>University of Michigan Cancer Center
>1500 E. Medical Center Drive
>7410 CCGC
>Ann Arbor MI 48109
>734-647-5623
>
>_______________________________________________
>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



More information about the Bioconductor mailing list