The edgeR update from 3.28.0 to 3.28.1 included some more sanity checks of library.size on norm.factors columns. Now many of our unit tests throw warnings about fishy norm.factors: - [ ] `test-as.FacileDataSet.R:27: warning: exampleFacileDataSet -> DGELists -> as.FacileDataSet norm factors don't multiply to 1` - [ ] `test-assay-normalization.R:18: warning: Normalization of rnaseq data is equivalent to edgeR::cpm norm factors don't multiply to 1` - [ ] `test-bioc-assay-containers.R:(18, 45, 46, 65): warning: fetch_assay_data results converted to DGEList norm factors don't multiply to 1`