hi Steve,
Thanks for pointing me to this bug. I have fixed this in version 1.0.9. So
now rlogTransformation should accept the given dds just like
varianceStabilizingTransformation.
To avoid making a copy of the dds but still re-estimate dispersions, one
can do:
design(dds) <- ~ 1
mcols(dds)$dispersion <- NULL
then either rlogTransformation or varianceStabilizingTransformation, which
will call estimateDispersions internally.
Mike
On Sun, Apr 21, 2013 at 8:13 PM, Steve Lianoglou <
mailinglist.honey...@gmail.com> wrote:
> Hi Michael,
>
> On Sat, Apr 20, 2013 at 12:18 AM, Michael Love
> <michaelisaiahl...@gmail.com> wrote:
> > hi Steve,
> >
> > Thanks. Can you send a minimal example of the error? I can provide an
> > alternate formula without error:
> >
> >> dds <- makeExampleDESeqDataSet()
> >> design(dds) <- formula(~ condition)
> >> rld <- rlogTransformation(dds)
>
> I'll see your `condition` and raise you a `treatment` :-)
>
> =======================
> R> set.seed(123)
> R> library(DESeq2)
> R> dds <- makeExampleDESeqDataSet()
> R> colData(dds)$treatment <- factor(sample(c("x", "y"), ncol(dds),
> replace=TRUE))
> R> design(dds) <- formula(~ condition + treatment)
>
> ## Fails for rlogTransformation
> R> dds.rlog <- rlogTransformation(dds)
> gene-wise dispersion estimates
> mean-dispersion relationship
> final dispersion estimates
> Error in SummarizedExperiment(assays = rlogData(object, samplesVector, :
> error in evaluating the argument 'assays' in selecting a method for
> function 'SummarizedExperiment': Error in
> modelMatrixNames[match(convertNames$from, modelMatrixNames)] <-
> convertNames$to :
> NAs are not allowed in subscripted assignments
>
> ## vst seems OK
> R> dds.vst <- varianceStabilizingTransformation(dds)
> gene-wise dispersion estimates
> mean-dispersion relationship
> final dispersion estimates
> ========================
>
> I'm using DESeq2_1.0.7 on R-3.0.0
>
> Thanks,
> -steve
>
> --
> Steve Lianoglou
> Computational Biologist
> Department of Bioinformatics and Computational Biology
> Genentech
>
[[alternative HTML version deleted]]
_______________________________________________
Bioc-devel@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/bioc-devel
