[BioC] interactions between variables in DEXSeq

[Alejandro Reyes]

Dear Elena,

Thank you for your email. If I understood correctly, to test for an
interaction between condition and type with the exons, it would be like
this (example in the pasilla dataset):

data("pasillaExons", package="pasilla")
pasillaExons<- estimateSizeFactors( pasillaExons )
pasillaExons<- estimateDispersions( pasillaExons, count ~ sample +
(condition * type) * exon )
formula0<- count ~ sample + exon + condition + type + condition:type
formula1<- count ~ sample + exon + condition + type + (condition:type)
* I(exon == exonID)
pasillaExons<- testForDEU( pasillaExons, formula0=formula0,
formula1=formula1 )


Best wishes,
Alejandro Reyes

>  Hello again,
>
>  I'm writing this time to ask about setting up more complex formulae to
>  test for significant interaction between independent variables in
>  DEXSeq. Using the example from the manual, to test for an interaction
>  between library type and condition, how would I set this up? The
>  syntax here is a bit more involved than with DESeq, and I can't seem
>  to find anything in the archives that answers my question...
>
>  Thanks,
>  Elena
>
>  _______________________________________________
>  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