[BioC] multicore and DEXSeq

Thu Aug 2 02:21:46 CEST 2012

It looks like I can't reproduce it either.  I just tried again with a
clean session and it seems to go through.  Thanks for your help!
Lynn

On 8/1/12 5:45 AM, "Alejandro Reyes" <alejandro.reyes at embl.de> wrote:

>Dear Lynn,
>
>I could not reproduce your error with my datasets.  Could you send me
>your ExonCountSet object, to have a closer look? With the code that you
>used and the sessionInfo() output?
>
>Alejandro
>
>> Alejandro,
>> No, I do not get the error using just one core.  When I run the same
>>exact
>> script, changing nCores=4 to nCores=1, there is no error message.
>>Thanks
>> for looking into this.
>> Lynn
>>
>>
>> On 7/31/12 10:09 AM, "Alejandro Reyes" <alejandro.reyes at embl.de> wrote:
>>
>>> Dear Lynn Amon, Dear Dan,
>>>
>>> Thanks for noticing the mistake in the vignette I will update it.
>>>
>>> Lynn, do you have the same error when you run it using just one core?
>>> If so, could you send me that error message?
>>>
>>> Alejandro
>>>
>>>
>>>> Actually, I'm still running into problems with the parallel package.
>>>> Here is my error with estimateDispersions:
>>>>
>>>>> glueExons <- estimateDispersions(glueExons,nCores=4)
>>>> Estimating Cox-Reid exon dispersion estimates using 4 cores. (Progress
>>>> report: one dot per 100 genes)
>>>> Error in function (classes, fdef, mtable)  :
>>>>     unable to find an inherited method for function "fData", for
>>>> signature "character"
>>>> In addition: Warning message:
>>>> In parallel::mclapply(allecs, FUN = funtoapply, mc.cores = mc.cores) :
>>>>     all scheduled cores encountered errors in user code
>>>>
>>>>
>>>> ________________________________________
>>>> From: Dan Tenenbaum [dtenenba at fhcrc.org]
>>>> Sent: Monday, July 30, 2012 1:02 PM
>>>> To: Lynn Amon
>>>> Cc: Lynn Amon [guest]; bioconductor at r-project.org; DEXSeq Maintainer
>>>> Subject: Re: [BioC] multicore and DEXSeq
>>>>
>>>> On Mon, Jul 30, 2012 at 12:59 PM, Lynn Amon
>>>> <lynn.amon at seattlebiomed.org> wrote:
>>>>> Dan,
>>>>> Thanks for the reply.  library(parallel) does work.  I used multicore
>>>>> because that what vignette for DEXSeq uses.
>>>>
>>>> Thanks for the clarification. The DEXSeq vignette should be
>>>> updated...Alejandro, would you mind doing that?
>>>>
>>>> Thanksc,
>>>> Dan
>>>>
>>>>
>>>>> Lynn
>>>>>
>>>>> On 7/30/12 12:55 PM, "Dan Tenenbaum" <dtenenba at fhcrc.org> wrote:
>>>>>
>>>>>> Hi Lynn,
>>>>>>
>>>>>> On Mon, Jul 30, 2012 at 12:51 PM, Lynn Amon [guest]
>>>>>> <guest at bioconductor.org> wrote:
>>>>>>>
>>>>>>> Hi,
>>>>>>> I am having trouble getting fitDespersions to recognize that I have
>>>>>>> multicore installed. I get an error from divideWork telling me to
>>>>>>> load
>>>>>>> parallel package but I have multicore and it detects 8 cores.  Here
>>>>>>> is
>>>>>>> some of my script to make an exonCountSet object and then estimate
>>>>>>> dispersions using four cores.
>>>>>>> Thanks,
>>>>>>> Lynn
>>>>>>>
>>>>>>>> glueExons <-
>>>>>>>>
>>>>>>>> 
>>>>>>>>read.HTSeqCounts(countfiles=file.path(bamPath,paste(rownames(sample
>>>>>>>>s)
>>>>>>>> ,"/
>>>>>>>>
>>>>>>>> 
>>>>>>>>htseq_exons.txt",sep="")),design=samples,flattenedfile=annotationFi
>>>>>>>>le
>>>>>>>> )
>>>>>>>>
>>>>>>>> sampleNames(glueExons) <- rownames(samples)
>>>>>>>>
>>>>>>>> expdata <- new("MIAME",name="GlueGenomics")
>>>>>>>> experimentData(glueExons) <- expdata
>>>>>>>>
>>>>>>>> library(DESeq)
>>>>>>>> genetable <- geneCountTable(glueExons)
>>>>>>>> glue_genes <- newCountDataSet(genetable,conditions=samples)
>>>>>>>> experimentData(glue_genes) <- expdata
>>>>>>>>
>>>>>>>> glueExons <- estimateSizeFactors(glueExons)
>>>>>>>> sizeFactors(glueExons)
>>>>>>>       WT_RY_0_12A    WT_RY_P3_12A   WT_RY_PIC_12A    WT_RY_PP_12A
>>>>>>> IL10_RZ_0_12A
>>>>>>>         1.0143924       0.8163463       1.2448730       0.4565408
>>>>>>> 2.7520441
>>>>>>>    IL10_RY_P3_12A IL10_RY_PIC_12A  IL10_RY_PP_12A
>>>>>>>         1.1597040       0.8003611       0.9501160
>>>>>>>>
>>>>>>>> library(multicore)
>>>>>>>> multicore:::detectCores()
>>>>>>> [1] 8
>>>>>>>> glueExons <- estimateDispersions(glueExons,nCores=4)
>>>>>>> Estimating Cox-Reid exon dispersion estimates using 4 cores.
>>>>>>> (Progress
>>>>>>> report: one dot per 100 genes)
>>>>>>> Error in divideWork(cds, funtoapply = toapply, fattr =
>>>>>>> "dispBeforeSharing",  :
>>>>>>>     Please load first parallel package or set parameter nCores to
>>>>>>>1...
>>>>>>>
>>>>>>>
>>>>>>
>>>>>> Load the "parallel" package, not the "multicore" package.
>>>>>> Dan
>>>>>>
>>>>>>
>>>>>>
>>>>>>>
>>>>>>>    -- output of sessionInfo():
>>>>>>>
>>>>>>> R version 2.15.0 (2012-03-30)
>>>>>>> Platform: x86_64-unknown-linux-gnu (64-bit)
>>>>>>>
>>>>>>> locale:
>>>>>>>    [1] LC_CTYPE=en_US.UTF-8       LC_NUMERIC=C
>>>>>>>    [3] LC_TIME=en_US.UTF-8        LC_COLLATE=en_US.UTF-8
>>>>>>>    [5] LC_MONETARY=en_US.UTF-8    LC_MESSAGES=en_US.UTF-8
>>>>>>>    [7] LC_PAPER=C                 LC_NAME=C
>>>>>>>    [9] LC_ADDRESS=C               LC_TELEPHONE=C
>>>>>>> [11] LC_MEASUREMENT=en_US.UTF-8 LC_IDENTIFICATION=C
>>>>>>>
>>>>>>> attached base packages:
>>>>>>> [1] stats     graphics  grDevices utils     datasets  methods
>>>>>>>base
>>>>>>>
>>>>>>> other attached packages:
>>>>>>> [1] DESeq_1.9.10       lattice_0.20-6     locfit_1.5-8
>>>>>>> DEXSeq_1.3.14
>>>>>>> [5] Biobase_2.17.6     BiocGenerics_0.3.0 multicore_0.1-7
>>>>>>>
>>>>>>> loaded via a namespace (and not attached):
>>>>>>>    [1] annotate_1.35.3       AnnotationDbi_1.19.28 biomaRt_2.13.2
>>>>>>>    [4] DBI_0.2-5             genefilter_1.39.0
>>>>>>>geneplotter_1.35.0
>>>>>>>    [7] grid_2.15.0           hwriter_1.3           IRanges_1.15.24
>>>>>>> [10] plyr_1.7.1            RColorBrewer_1.0-5    RCurl_1.91-1
>>>>>>> [13] RSQLite_0.11.1        splines_2.15.0        statmod_1.4.14
>>>>>>> [16] stats4_2.15.0         stringr_0.6.1         survival_2.36-14
>>>>>>> [19] XML_3.9-4             xtable_1.7-0
>>>>>>>
>>>>>>>
>>>>>>> --
>>>>>>> Sent via the guest posting facility at bioconductor.org.
>>>>>>>
>>>>>>> _______________________________________________
>>>>>>> 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
>>>>>
>>>
>>>
>>
>
>