[BioC] Multiple contrasts in DESeq2

Hugo Varet hugo.varet at pasteur.fr
Mon Dec 23 11:10:40 CET 2013

Dear Michael Love and list members,

I'm used to analyze RNA-seq data with your nice DESeq2 package (1.2.0) 
and I sometimes have to test complex null hypotheses using contrasts. 
For one of my project, I would like using multiple contrasts, i.e. 
testing for example (H0: beta1-beta2 = beta3-beta4 = 0) against (H1: 
beta1-beta2 != 0 and/or beta3-beta4 != 0), where the beta's are the 
regression coefficients of the models. To do so, I would like using a 
numeric matrix for the contrast argument in results(), but it only 
accepts a numeric vector.

Hence, I was wondering if you planned to extend DESeq2 to the use of 
multiple contrasts?

Best regards,


Hugo Varet
Plate-Forme Transcriptome et Epigénome
Institut Pasteur

More information about the Bioconductor mailing list