[BioC] DEXSeq design formula

Alejandro Reyes alejandro.reyes at embl.de
Tue May 13 10:55:42 CEST 2014


Hi Simone,

Yes, this is correct. And then you pass the reduced model to testForDEU.
I will double check the documentation and make sure its clearer.

Bests,
Alejandro Reyes

> Hello!
>
> Just a quick question regarding DEXSeq:
>
> I have more variables in my experiment (9 samples), i.e. 3 different
> celltypes (variable of interest) taken from 4 different patients. To
> take into account when samples come from the same patient I included
> an additional factor to the sampleData table.
>
> I understand that I have to use the following formulae when it comes
> to testing for differential exon usage:
>
> formulaFullModel = ~ sample + exon + patient:exon + celltype:exon
> formulaReducedModel = ~ sample + exon + patient:exon
>
> And I assume (but  it's not quite clear to me from reading the
> vignette) that I also have to add the additional patient variable to
> the design already when creating the DEXSeqDataSetFromHTSeq:
>
> dxd = DEXSeqDataSetFromHTSeq(
>    countFiles,
>    sampleData=sampleTable,
>    design= ~ sample + exon + patient:exon + celltype:exon,
>    flattenedfile=flattenedFile)
>
> Is this correct?
>
> Best,
> Simone
>
> _______________________________________________
> 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



More information about the Bioconductor mailing list