[BioC] Sudden Issue with add for lymphGate in flowCore

[Martin Morgan]

On 11/13/2011 12:08 PM, Aric Gregson wrote:
> Hello,
>
> I have an odd issue that I have been unable thus far to resolve.
> Literally overnight it has become impossible to add a lymphGate filter
> to a workFlow. This is specific for the lymphGate filter only, as any
> other type of filter works just fine. The error that I am receiving is
> shown after the Sweave output here. It happens regardless of whether
> the code is run in Sweave or manually in R.
>
> [snip]
> 10 : echo keep.source term verbatim (label =
> createWorkFlowBoundaryTransformLogicle)
> 11 : echo keep.source term verbatim (label = LiveDeadUVFilterAlt)
> 12 : keep.source term tex (label = summaryLive)
> 13 : echo keep.source term verbatim pdf (label = plotLiveDead)
> 14 : keep.source (label = LiveDeadUVFiltercd3)
> 15 : keep.source (label = plotLiveDeadcd3)
> 16 : echo keep.source term verbatim (label = singletfilterCreate)
> 17 : echo keep.source term verbatim pdf (label = plotSingletfilter)
> 18 : echo keep.source term verbatim (label = lymphocyteFilterCD4CD3)
> 19 : echo keep.source term verbatim (label = add.lymph.Gate)
>
> Error:  chunk 19 (label = add.lymph.Gate)
> Error in function (classes, fdef, mtable)  :
>    unable to find an inherited method for function "add", for signature
> "workFlow", "list"
> In addition: Warning messages:
> 1: In readBin(con, dattype, n = (offsets["dataend"] -
> offsets["datastart"] +  :
>    'signed = FALSE' is only valid for integers of sizes 1 and 2
> 2: In readBin(con, dattype, n = (offsets["dataend"] -
> offsets["datastart"] +  :
>    'signed = FALSE' is only valid for integers of sizes 1 and 2
> 3: In readBin(con, dattype, n = (offsets["dataend"] -
> offsets["datastart"] +  :
>    'signed = FALSE' is only valid for integers of sizes 1 and 2
> 4: In readBin(con, dattype, n = (offsets["dataend"] -
> offsets["datastart"] +  :
>    'signed = FALSE' is only valid for integers of sizes 1 and 2
> 5: In readBin(con, dattype, n = (offsets["dataend"] -
> offsets["datastart"] +  :
>    'signed = FALSE' is only valid for integers of sizes 1 and 2
> 6: In readBin(con, dattype, n = (offsets["dataend"] -
> offsets["datastart"] +  :
>    'signed = FALSE' is only valid for integers of sizes 1 and 2
> 7: In readBin(con, dattype, n = (offsets["dataend"] -
> offsets["datastart"] +  :
>    'signed = FALSE' is only valid for integers of sizes 1 and 2
>
>
> The only difference overnight is that I 'upgraded' the tcl/tk package
> on my computer. Since then I have reinstalled R and ran
> upgrade.packages for both R and Bioconductor, but the problem persists.
> I suspect I have not provided enough information here. Any ideas on how
> to figure this out would be greatly appreciated.

Hi Aric -- instead of running 'Sweave' on your vignette, trying running 
'Stangle'. This produces an R script file that you can step through. 
Likely you'll get to the code corresponding to chunk 19, and see the 
error. Likely you can narrow this down to a single function call that 
causes the problem and, with luck, identify a parameter to a call to 
add() (probably the second parameter) that is a 'list' rather than what 
it is supposed to be. But at that point you might be able to give us a 
simple-to-reproduce example. Martin

>
> Thanks, Aric
> -----------------
>
>   sessionInfo()
> R version 2.13.2 (2011-09-30)
> Platform: amd64-portbld-freebsd9.0 (64-bit)
>
> locale:
> [1] C
>
> attached base packages:
>   [1] splines   tools     grid      stats     graphics  grDevices
> utils
>   [8] datasets  methods   base
>
> other attached packages:
>   [1] xtable_1.6-0          Rgraphviz_1.30.1      flowQ_1.12.0
>   [4] latticeExtra_0.6-19   RColorBrewer_1.0-5    parody_1.10.0
>   [7] bioDist_1.24.1        KernSmooth_2.23-7     outliers_0.14
> [10] flowStats_1.13.0      cluster_1.14.1        mvoutlier_1.9.4
> [13] robCompositions_1.5.0 car_2.0-11            survival_2.36-10
> [16] nnet_7.3-1            compositions_1.10-2   energy_1.4-0
> [19] MASS_7.3-14           boot_1.3-2            tensorA_0.36
> [22] rgl_0.92.798          fda_2.2.7             zoo_1.7-6
> [25] flowWorkspace_0.99.22 IDPmisc_1.1.16        flowViz_1.16.0
> [28] lattice_0.19-33       flowCore_1.18.0       rrcov_1.3-01
> [31] pcaPP_1.9-44          mvtnorm_0.9-9991      robustbase_0.7-8
> [34] Biobase_2.12.2        XML_3.4-3             RBGL_1.29.0
> [37] graph_1.30.0          multicore_0.1-7       Rmpi_0.5-9
> [40] fortunes_1.4-2
>
> loaded via a namespace (and not attached):
> [1] AnnotationDbi_1.14.1 DBI_0.2-5            RSQLite_0.10.0
> [4] annotate_1.30.1      feature_1.2.8        geneplotter_1.30.0
> [7] ks_1.8.4             stats4_2.13.2
>
> _______________________________________________
> 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


-- 
Computational Biology
Fred Hutchinson Cancer Research Center
1100 Fairview Ave. N. PO Box 19024 Seattle, WA 98109

Location: M1-B861
Telephone: 206 667-2793