[BioC] arrayQualityMetrics NULL XML-node Error; Anyone still getting this?
Wolfgang Huber
whuber at embl.de
Sun Oct 9 20:11:41 CEST 2011
Dear Eric
thank you. I understand your general concern about stability in a
production environment, but specifically here this is not a real problem:
- Browser: older, released versions of Firefox (e.g. 7.0.1) or Chrome
(and probably others that are HTML5-capable, I have not tried) will be
just as OK.
- R and Bioconductor: you will then have to wait until the next release,
at the end of this month / early November.
- cairo and libxml2, actually the version numbers that we exchanged
previously were irrelevant. What is needed it the output of
$ pkg-config --modversion cairo libxml-2.0
1.10.2
2.7.8
What do you get here?
cairo version 1.10.2 and libxml-2.0 version 2.7.8 were both released in
2010, almost a year ago, see http://cairographics.org and
http://xmlsoft.org/news.html
Best wishes
Wolfgang
Oct/9/11 5:31 AM, Paniagua, Eric scripsit::
> Dear Wolfgang,
>
> Thanks for taking a look and for your quick response. I will look at the potential upgrades you are suggesting, but I see that several key versions you are using are unstable or development versions including R, your browser, and AQM itself.
>
> I need to produce software for a production system rather than personal use, so I hesitate to introduce dev/unstable dependencies. Do you know offhand the most recent combination of stable versions that work together correctly, for instance with Bioconductor release 2.8?
>
> Thanks,
> Eric
> ________________________________________
> From: bioconductor-bounces at r-project.org [bioconductor-bounces at r-project.org] on behalf of Wolfgang Huber [whuber at embl.de]
> Sent: Saturday, October 08, 2011 9:44 AM
> To: bioconductor at r-project.org
> Subject: Re: [BioC] arrayQualityMetrics NULL XML-node Error; Anyone still getting this?
>
> Dear Eric
>
> thank you for sending a reproducible example. I have run the same code
> as you, and this produced a report without error messages and with all
> elements that are supposed to be interactive (Figs. 3, 6, 12 and the
> table) behaving as expected. The code and the report are here:
> http://www-huber.embl.de/users/whuber/bioc-list/111008/
>
> I have the following system parameters and session info as below, so my
> recommendation is for you to update to more recent versions. More
> comments see below.
>
> $ pkg-config --version cairo
> 0.26
> $ xml2-config --version
> 2.7.8
> # arrayQualityMetrics version: 3.9.5.
> # Firefox 8.0
>
> > sessionInfo()
> R Under development (unstable) (2011-10-07 r57184)
> Platform: x86_64-apple-darwin10.8.0/x86_64 (64-bit)
>
> locale:
> [1] C
>
> attached base packages:
> [1] stats graphics grDevices utils datasets methods base
>
> other attached packages:
> [1] hgu133acdf_2.8.0 ALLMLL_1.2.8
> [3] affy_1.31.6 Biobase_2.13.10
> [5] arrayQualityMetrics_3.9.5 fortunes_1.4-2
>
> loaded via a namespace (and not attached):
> [1] AnnotationDbi_1.15.29 BiocInstaller_1.1.28 Biostrings_2.21.11
> [4] Cairo_1.5-0 DBI_0.2-5 Hmisc_3.8-3
> [7] IRanges_1.11.27 KernSmooth_2.23-6 RColorBrewer_1.0-5
> [10] RSQLite_0.10.0 SVGAnnotation_0.9-0 XML_3.4-3
> [13] affyPLM_1.29.3 affyio_1.21.2 annotate_1.31.1
> [16] beadarray_2.3.10 cluster_1.14.0 genefilter_1.35.0
> [19] grid_2.15.0 hwriter_1.3 lattice_0.19-33
> [22] latticeExtra_0.6-18 limma_3.9.21 preprocessCore_1.15.0
> [25] reshape_0.8.4 setRNG_2009.11-1 splines_2.15.0
> [28] survival_2.36-10 tools_2.15.0 vsn_3.21.2
> [31] xtable_1.5-6 zlibbioc_0.1.8
>
> ------------------------------------------------
> The fact that for you Figs. 6 and 12 'work' -these are line plots- and
> that Fig. 3 does not -this is a scatter plot- suggests that what you are
> encountering is an aspect of the brittleness in the way that
> arrayQualityMetrics (via SVGAnnotation) interacts with SVG files
> produced by a third party (cairo). Essentially, that interaction is not
> based on a defined protocol, but on unilateral reverse engineering.
>
> SVGAnnotation is a fantastic piece of 'proof of concept' software, but
> the situation is, of course, not satisfatory for applications that are
> supposed to be used by many different people and on different platforms.
> As soon as I can find the time, I will see how it can improved ( e.g.
> gridSVG, where we have more control over the SVG ).
>
> Best wishes
> Wolfgang
>
>
> Oct/7/11 8:59 PM, Paniagua, Eric scripsit::
>> Hi Bioconductors (hopefully still including Wolfgang),
>>
>> I have updated my packages, as per this list's posting guide, but to no avail. I have also searched the archives, and I discovered this thread where someone had a very similar problem (http://thread.gmane.org/gmane.science.biology.informatics.conductor/31622/focus=31666). I do not see there any solution. It's been about a year since that thread circulated, so I was wondering if anyone else has encountered this XML problem when running arrayQualityMetrics.
>>
>> It is not clear that my error is exactly the same as what James (from the above thread) encountered, but I'm thinking it's likely related. Maybe there is a parallel code block to the one throwing an error for James that wasn't fixed and is only now being observed. The differences from James's report follow:
>>
>> 1) I do get sections 4 and 5 in the index.html report file.
>>
>> 2) I have a notice beneath the caption for Figure 3 which reads: "Note: the figure is static - enhancement with interactive effects failed. This is either due to a version incompatibility of the 'SVGAnnotation' R package and your version of 'Cairo' or 'libcairo', or due to plot misformating. Please consult the Bioconductor mailing list, or contact the maintainer of 'arrayQualityMetrics' with a reproducible example in order to fix this problem."
>>
>> $ pkg-config --version cairo
>> 0.21
>>
>> $ xml2-config --version
>> 2.6.26
>>
>> The binary libcairo file installed is at /usr/lib64/libcairo.so.2.9.2
>>
>> Figures 6 and 12 *are* interactive, which leads me to think it could be either a version mismatch issue between libcairo and SVGAnnotation or a code issue in the arrayQualityMetrics package itself.
>>
>> I also have another case with the same error manifesting slightly differently, but I figure it's best to have everything working correctly for sample code from the vignette first (http://www.bioconductor.org/packages/2.8/bioc/vignettes/arrayQualityMetrics/inst/doc/arrayQualityMetrics.pdf and below). I notice that some of my versions are slightly newer than those from the vignette.
>>
>> Details:
>> * R was run as: R --vanilla
>> * R and all packages are installed on a 48-core server, 64-bit, running CentOS 5.5 (final)
>> * For viewing, I copied the output to my local machine (dual-core laptop running Ubuntu 10.10)
>> * I viewed index.html in Google Chrome 14.0.835.186
>>
>> My example R session demonstrating the error follows.
>>
>> I would greatly appreciate any feedback, suggestions, questions, and/or answers you may be able to provide.
>>
>> Thanks,
>> Eric Paniagua
>>
>> PS: I apologize for the lack of monospace font in what follows; I can't find how to set it in the OWA client.
>>
>> R version 2.13.1 (2011-07-08)
>> Copyright (C) 2011 The R Foundation for Statistical Computing
>> ISBN 3-900051-07-0
>> Platform: x86_64-unknown-linux-gnu (64-bit)
>>
>> R is free software and comes with ABSOLUTELY NO WARRANTY.
>> You are welcome to redistribute it under certain conditions.
>> Type 'license()' or 'licence()' for distribution details.
>>
>> R is a collaborative project with many contributors.
>> Type 'contributors()' for more information and
>> 'citation()' on how to cite R or R packages in publications.
>>
>> Type 'demo()' for some demos, 'help()' for on-line help, or
>> 'help.start()' for an HTML browser interface to help.
>> Type 'q()' to quit R.
>>
>>> library("ALLMLL")
>> Loading required package: affy
>> Loading required package: Biobase
>>
>> Welcome to Bioconductor
>>
>> Vignettes contain introductory material. To view, type
>> 'browseVignettes()'. To cite Bioconductor, see
>> 'citation("Biobase")' and for packages 'citation("pkgname")'.
>>
>>> data("MLL.A")
>>> library("arrayQualityMetrics")
>>> arrayQualityMetrics(expressionset = MLL.A[, 1:5],
>> + outdir = "Report_for_MLL_A",
>> + force = TRUE,
>> + do.logtransform = TRUE)
>> The directory 'Report_for_MLL_A' has been created.
>> Error in UseMethod("xmlAttrs", node) :
>> no applicable method for 'xmlAttrs' applied to an object of class "NULL"
>> KernSmooth 2.23 loaded
>> Copyright M. P. Wand 1997-2009
>>> traceback()
>> No traceback available
>>> warnings()
>> NULL
>>> sessionInfo()
>> R version 2.13.1 (2011-07-08)
>> Platform: x86_64-unknown-linux-gnu (64-bit)
>>
>> locale:
>> [1] C
>>
>> attached base packages:
>> [1] stats graphics grDevices utils datasets methods base
>>
>> other attached packages:
>> [1] hgu133acdf_2.8.0 arrayQualityMetrics_3.8.0
>> [3] ALLMLL_1.2.8 affy_1.30.0
>> [5] Biobase_2.12.2
>>
>> loaded via a namespace (and not attached):
>> [1] AnnotationDbi_1.14.1 Biostrings_2.20.4 Cairo_1.5-0
>> [4] DBI_0.2-5 Hmisc_3.8-3 IRanges_1.10.6
>> [7] KernSmooth_2.23-6 RColorBrewer_1.0-5 RSQLite_0.10.0
>> [10] SVGAnnotation_0.9-0 XML_3.4-3 affyPLM_1.28.5
>> [13] affyio_1.20.0 annotate_1.30.1 beadarray_2.2.0
>> [16] cluster_1.14.0 genefilter_1.34.0 grid_2.13.1
>> [19] hwriter_1.3 lattice_0.19-33 latticeExtra_0.6-18
>> [22] limma_3.8.3 preprocessCore_1.14.0 setRNG_2009.11-1
>> [25] splines_2.13.1 survival_2.36-10 tools_2.13.1
>> [28] vsn_3.20.0 xtable_1.5-6
>> _______________________________________________
>> Bioconductor mailing list
>> Bioconductor at r-project.org
>> https://stat.ethz.ch/mailman/listinfo/bioconductor
>> Search the archives: http://news.gmane.org/gmane.science.biology.informatics.conductor
>
>
> --
>
>
> Wolfgang Huber
> EMBL
> http://www.embl.de/research/units/genome_biology/huber
>
> _______________________________________________
> Bioconductor mailing list
> Bioconductor at r-project.org
> https://stat.ethz.ch/mailman/listinfo/bioconductor
> Search the archives: http://news.gmane.org/gmane.science.biology.informatics.conductor
--
Wolfgang Huber
EMBL
http://www.embl.de/research/units/genome_biology/huber
More information about the Bioconductor
mailing list