Hi everyone,

I’m using DESeq2 to analyze RNA-Seq data. I have basically 2 factors: Genotype and Treatment. I build the design matrix as ~ Genotype * Treatment to capture interaction effects (Genotype:Treatment) as well. After correctly releveling the factors, I get the coefficients that I want. So far so good.

The question is, is there any way to test if the interaction term itself is significant? In more general terms, is there any way to create something like an ANOVA table on the DESeqResults object?

Cheers,

Cihan

Yes you can test the interaction term, see examples in ?results.

No, we do not build a series of results tables removing variables one by one from the end of the design.