[BioC] parallel and GSVA

[Robert Castelo]

hi David,

ok, then it's an issue with the parallel package in this platform. you 
can put parallel.sz=8 but that should be in fact the default behavior 
raising the error. GSVA was trying to use all your processors by default 
and that caused the error.

i work usually on linux and therefore i'd need some time to investigate 
this issue in mac, maybe you can try Steve's suggestion to work on the 
terminal window of mac? does it work that way?

robert.

On 01/11/2013 02:26 PM, David Iles wrote:
> Hi Robert,
>
> Thanks for your prompt response.
>
> Yes, setting parallel.sz=1 does work. Thanks for that. I have 8 cores available, so can I set parallel.sz=8?
>
> Thanks
>
> Dave
>
> Dr David Iles
> School of Biology
> University of Leeds
> Leeds LS2 9JT
>
> d.e.iles at leeds.ac.uk
>
>
>
>
>
> On 11 Jan 2013, at 13:11, Robert Castelo<robert.castelo at upf.edu>  wrote:
>
>> Dear David,
>>
>> just to discard any other issue not related to the parallel execution in
>> your system, could you try to call the gsva() function adding the
>> argument parallel.sz=1? does it work?
>>
>> cheers,
>> robert.
>>
>> On 01/11/2013 01:18 PM, David Iles wrote:
>>> Hi Folks,
>>>
>>> I want to run a GSVA analysis on a set of Affymetrix hgu133plus2 arrays. To familiarise myself with the GSVA code, I worked through the example in the vignette, but encountered the following issue...
>>>
>>>> data(leukemia)
>>>> data(c2BroadSets)
>>>> library(parallel)
>>>> filtered_eset<- nsFilter(leukemia_eset, require.entrez=TRUE, remove.dupEntrez=TRUE,
>>> +                           var.func=IQR, var.filter=TRUE, var.cutoff=0.5, filterByQuantile=TRUE,
>>> +                           feature.exclude="^AFFX")
>>>> leukemia_filtered_eset<- filtered_eset$eset
>>>> leukemia_es<- gsva(leukemia_filtered_eset, c2BroadSets, min.sz=10, max.sz=500, verbose=TRUE)$es.obs
>>> Mapping identifiers between gene sets and feature names
>>>    |                                                                                                               |   0%)
>>>
>>> The process has forked and you cannot use this CoreFoundation functionality safely. You MUST exec().
>>> Break on __THE_PROCESS_HAS_FORKED_AND_YOU_CANNOT_USE_THIS_COREFOUNDATION_FUNCTIONALITY___YOU_MUST_EXEC__() to debug.
>>>
>>> ....etcetcetc. The process hangs here.
>>>
>>> I get the same messages with my own data. I must have overlooked something. Any suggestions?
>>>
>>> Dr David Iles
>>> School of Biology
>>> University of Leeds
>>> Leeds LS2 9JT
>>>
>>> d.e.iles at leeds.ac.uk<mailto:d.e.iles at leeds.ac.uk>
>>>
>>>> sessionInfo()
>>> R version 2.15.2 (2012-10-26)
>>> Platform: x86_64-apple-darwin9.8.0/x86_64 (64-bit)
>>>
>>> locale:
>>> [1] en_GB.UTF-8/en_GB.UTF-8/en_GB.UTF-8/C/en_GB.UTF-8/en_GB.UTF-8
>>>
>>> attached base packages:
>>> [1] parallel  grid      stats     graphics  grDevices utils     datasets  methods   base
>>>
>>> other attached packages:
>>>   [1] RBGL_1.34.0          BiocInstaller_1.8.3  GSVAdata_0.99.9      hgu95a.db_2.8.0      org.Hs.eg.db_2.8.0
>>>   [6] RSQLite_0.11.2       DBI_0.2-5            GSVA_1.6.1           GSEABase_1.20.1      annotate_1.36.0
>>> [11] AnnotationDbi_1.20.3 Rgraphviz_2.2.1      graph_1.36.1         RColorBrewer_1.0-5   limma_3.14.3
>>> [16] genefilter_1.40.0    Biobase_2.18.0       BiocGenerics_0.4.0
>>>
>>> loaded via a namespace (and not attached):
>>> [1] IRanges_1.16.4  splines_2.15.2  stats4_2.15.2   survival_2.37-2 tools_2.15.2    XML_3.95-0.1    xtable_1.7-0
>>>
>>> _______________________________________________
>>> 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
>>>
>>
>> --
>> Robert Castelo, PhD
>> Associate Professor
>> Dept. of Experimental and Health Sciences
>> Universitat Pompeu Fabra (UPF)
>> Barcelona Biomedical Research Park (PRBB)
>> Dr Aiguader 88
>> E-08003 Barcelona, Spain
>> telf: +34.933.160.514
>> fax: +34.933.160.550
>>
>> _______________________________________________
>> 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
>
>

-- 
Robert Castelo, PhD
Associate Professor
Dept. of Experimental and Health Sciences
Universitat Pompeu Fabra (UPF)
Barcelona Biomedical Research Park (PRBB)
Dr Aiguader 88
E-08003 Barcelona, Spain
telf: +34.933.160.514
fax: +34.933.160.550