Hello,

I'm using DEseq2 on Illumina RNA-seq data. What I want to do is to create plots for my genes of interest, with the counts on the y-axis, and my sample types on the x-axis. This can be done very nicely with the function plotCounts(dds, gene=topGene, intgroup="timepoint", returnData=T) and subsequently calling the plot with ggplot2 . But now I would also like to add p-values showing the differences between my sample types. I tried using ggplot2 with geom_signif() and stat_compare_means() but even with the top gene from one of the results tables, I only get ns (non significant) plotted.

Is there an easy way to add significance levels to the output of plotCounts()?

Thanks for your help, Anne

This downstream plotting of DESeq2 output is up to you. I don't think it will be useful to try any of those geom's or stat's as they are for simple group testing, and for a number of reasons, the DESeq2 model is not doing the same inference as these.

So you're on your own. If you're new to ggplot2, they have a lot of online documentation, or you can consult someone in your group/department/institute who has more experience.