Hi,
I was wondering whether anybody has experience with equivalence testing using edgeR. Usually people are interested in differential expression, and that works out of the box with edgeR. But what if you are interested in genes that do not change in two conditions? Picking those with large p-value from standard difference tests won't work because absence of evidence is not evidence of absence. I know that for t-tests one can set up TOST tests of equivalence. In principle that should be doable also with the negative binomial test edgeR is built on. Has anybody tried that?
Thank you for your feedback.
Roberto
I'd recommend you ask this on either the bioconductor support site or the stats stackoverflow. The former is probably the best route, since the edgeR authors are quite responsive.