[BioC] farms package fails on estrogen data

[bmb at bmbolstad.com]

ReadAffy uses allocMatrix() to allocate the memory into which it intends
to place probe intensities read from the file. Based on information in the
CEL file header it determines grid dimensions to determine how many probe
intensities are expected. Exact parsing depends on the CEL file format of
which there are a few, but generally speaking it is expected that there is
an intensity to be read for every location in the grid. Zeros can arise in
truncated and corrupted CEL files. But many of those types of situations
should be detected and an error thrown. There may be other mechanisms for
generation of such probe intensities which make them valid.



> On 03/16/2010 08:25 AM, Javier Pérez Florido wrote:
>> Dear list,
>> I've tried to use l.farms (farms package)
>> http://www.bioinf.jku.at/software/farms/farms.html
>> for preprocessing the estrogen data set available at
>> http://www.bioconductor.org/packages/release/data/experiment/html/estrogen.html
>>
>>
>> When trying to preprocess using l.farms, the following error comes up:
>>
>> background correction: none
>> normalization: loess
>> PM/MM correction : pmonly
>> expression values: farms
>> background correcting...done.
>> normalizing...Error: NA/NaN/Inf in call to external function (arg 1)
>>
>> q.farms works for the same data set and l.farms works for the Dilution
>> data set, so, I don't know what is going on here.
>>
>> Any suggestions?
>
> Hi Javier --
>
> If I
>
>   library(farms)
>   library(estrogen)
>   setwd(system.file("extdata", package="estrogen"))
>   ab = ReadAffy()
>   l.farms(ab)
>
> I see the error you report. It is because
>
>> sum(intensity(ab) == 0)
> [1] 5
>
> so that log transformation generates -Inf and then the error. It doesn't
> seem too bad to
>
>   intensity(ab) = 1 + intensity(ab)
>
> [I don't know enough about the inner workings of affyio, but I'm
> surprised that I sometimes see
>
>> range(intensity(ReadAffy()))
> [1]     0 46131
>
>> range(intensity(ReadAffy))
> [1] 1.563578e-316  4.613100e+04
>
> suggesting either initialized memory or unanticipated 'empty' intensity
> grid coordinates)]
>
> Martin
>
>
>> Javier
>>
>> sessionInfo()
>> R version 2.10.1 (2009-12-14)
>> i386-pc-mingw32
>>
>> locale:
>> [1] LC_COLLATE=Spanish_Spain.1252  LC_CTYPE=Spanish_Spain.1252
>> [3] LC_MONETARY=Spanish_Spain.1252 LC_NUMERIC=C
>> [5] LC_TIME=Spanish_Spain.1252
>>
>> attached base packages:
>> [1] stats     graphics  grDevices utils     datasets  methods   base
>>
>> other attached packages:
>> [1] affyPLM_1.22.0       preprocessCore_1.8.0 gcrma_2.18.1
>> [4] affydata_1.11.10     farms_1.4.0          MASS_7.3-4
>> [7] hgu95av2cdf_2.5.0    affy_1.24.2          Biobase_2.6.1
>>
>> loaded via a namespace (and not attached):
>> [1] affyio_1.14.0      Biostrings_2.14.12 IRanges_1.4.11
>> splines_2.10.1
>> [5] tools_2.10.1
>>
>> _______________________________________________
>> 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
>
>
> --
> Martin Morgan
> Computational Biology / Fred Hutchinson Cancer Research Center
> 1100 Fairview Ave. N.
> PO Box 19024 Seattle, WA 98109
>
> Location: Arnold Building M1 B861
> Phone: (206) 667-2793
>
> _______________________________________________
> 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
>