[BioC] problem with aveLogCPM.default in edgeR

[Gordon K Smyth]

You can avoid the problem by forming a DGEList:

   y <- DGEList(counts=exprs(dataNormgcOff))
   y$offset <- -offst(dataNormgcOff)

Then

   y <- estimateGLMCommonDisp(y, design)

etc.

Gordon

> Date: Thu, 16 May 2013 15:57:53 -0400
> From: <suzy.stiegelmeyer at syngenta.com>
> To: <bioconductor at r-project.org>
> Subject: [BioC] problem with aveLogCPM.default in edgeR
>
> Hi,

> I recently upgraded edgeR from 3.0.8 to 3.2.3 and I'm noticing some 
> differences.  I have some data that I normalized with EDASeq.  I 
> attempted to calculate the trended dispersion and I get the following 
> error:
>
>> dglmtrend=estimateGLMTrendedDisp(exprs(dataNormgcOff),design,offset=-offst(dataNormgcOff))
>
> Error in t(y) + prior.count.scaled : non-conformable arrays
>
>> class(dataNormgcOff)
>
> [1] "SeqExpressionSet"
>
> attr(,"package")
>
> [1] "EDASeq"
>
>> dim(exprs(dataNormgcOff))
>
> [1] 19062    36
>
>> dim(offst(dataNormgcOff))
>
> [1] 19062    36
>
> The error seems to occur in the aveLogCPM.default function due to matrix 
> addition on two matrices with differing dimensions. Line 34 reads as:

> abundance <- mglmOneGroup(t(t(y)+prior.count.scaled),dispersion=dispersion,offset=offset)
>
> I no longer get an error if I change it to:

> abundance <- mglmOneGroup(y+prior.count.scaled,dispersion=dispersion,offset=offset)
>
> I don't think this is the best solution to fix all scenarios since I 
> don't know this code very well.  So, I see some things have changed and 
> I'm wondering if I need to make some changes in how I call the function 
> or if there is really a bug of some kind here.
>
> Thanks in advance for your help,
> Suzy
>
>> sessionInfo()
> R version 3.0.0 (2013-04-03)
> Platform: x86_64-w64-mingw32/x64 (64-bit)
>
> locale:
> [1] LC_COLLATE=English_United States.1252  LC_CTYPE=English_United States.1252
> [3] LC_MONETARY=English_United States.1252 LC_NUMERIC=C
> [5] LC_TIME=English_United States.1252
>
> attached base packages:
> [1] parallel  stats     graphics  grDevices utils     datasets  methods   base
>
> other attached packages:
> [1] EDASeq_1.6.0         aroma.light_1.30.1   matrixStats_0.8.1    ShortRead_1.18.0
> [5] latticeExtra_0.6-24  RColorBrewer_1.0-5   Rsamtools_1.12.3     lattice_0.20-15
> [9] Biostrings_2.28.0    GenomicRanges_1.12.3 IRanges_1.18.1       Biobase_2.20.0
> [13] BiocGenerics_0.6.0   edgeR_3.2.3          limma_3.16.3
>
> loaded via a namespace (and not attached):
> [1] annotate_1.38.0      AnnotationDbi_1.22.5 bitops_1.0-5         DBI_0.2-7
> [5] DESeq_1.12.0         genefilter_1.42.0    geneplotter_1.38.0   grid_3.0.0
> [9] hwriter_1.3          R.methodsS3_1.4.2    RSQLite_0.11.3       splines_3.0.0
> [13] stats4_3.0.0         survival_2.37-4      tools_3.0.0          XML_3.96-1.1
> [17] xtable_1.7-1         zlibbioc_1.6.0      _________________________________
>
> Suzy Stiegelmeyer, PhD
> Computational Biologist
> Bioinformatics
>
> Syngenta Biotechnology, Inc.
> 3054 Cornwallis Rd
> Research Triangle Park, NC
> 27709
> USA
>
> phone +1 919 281 7472
>
> suzy.stiegelmeyer at syngenta.com<mailto:suzy.stiegelmeyer at syngenta.com>
> www.syngenta.com<http://www.syngenta.com/>
>

______________________________________________________________________
The information in this email is confidential and intend...{{dropped:4}}