In my experiment, I have three groups of plants: wild-type (WT), induced (IN), and uninduced (UN). These groups have been established to explore the effects of a specific gene, which I refer to as gene DM, that I can induce with ethanol (EtOH) treatment. The IN group is composed of plants with the DM gene induced by 1% EtOH treatment. The UN group contains the same genetically modified plants as the IN group but without the induction, having received a mock treatment with water. Additionally, I have the WT group, which consists of non-transformed wild-type plants treated with EtOH to observe the effects of the EtOH treatment by itself.
This experiment also includes a time series component, with samples collected at multiple time points after treatment: 0, 1.5, 3, 6, 12, and 24 hours post-induction (hpi).
My question centers on how to analyze the RNA-seq data to identify the differentially expressed genes (DEGs) that result from activating the DM gene in the IN group compared to the UN group. Crucially, my analysis needs to account for the effects of EtOH treatment, which are present in the WT group, to ensure that the DEGs I identify are specific to the DM gene's function and not a response to EtOH. My challenge is to design a statistical analysis that effectively separates the influence of the DM gene induction from the EtOH treatment effects.
In DESeq2, I can use ~ Genotype + Treatment + Time + Genotype:Treatment + Time:Treatment + Genotype:Time + Genotype:Treatment:Time
but how to get contrast to study IN vs UN, and remove the effect of EtOH treatment caused in WT? I am not familiar with edgeR, does it support more complicated design?