Michael Love

Re: Swish, it's fairly similar to running DESeq2 or other Bioc methods: ``` y

Just to check: do you have length bias in your data (are counts roughly proportional to effective transcript length)?

The standard is the code chunk in the vignette: ``` txi

To keep the code simpler: ``` txi

No difference. I only prefer people use `...Tximport()` only because we had some people using `txi$counts` alone and not using the `countsFromAbundance` argument, and then calling that the "tximport" method,...

It seems this is unambiguous from the documentation, right? Is there any question left about what to do? I can’t think how to explain it in sentences that are different...

TPM is an abundance measure and becomes `"abundance"` in the list of matrices. NumReads is an estimated count and becomes `"counts"` in the list of matrices. Each software has different...

These links won't work bc you left a space between [] and (), can you address?

Right, as in your discussion with Susan Holmes https://twitter.com/krassowski_m/status/1230896371034705925 Maybe it's time to add a category for "Methods with structure on features", so that would include pathway enrichment as well...

Thanks @frederikziebell and @stephens999 I'll look into adding this to DESeq2 in this devel cycle.