exprs() returning NULL about scater HOT 2 CLOSED

exprs() itself is not deprecated, but now points to the "logcounts" assay, as this was the typical use.

In this specific case, we still store the log-transformed "exprs" in the output, but exprs() doesn't point there anymore. To get at it, you'll need to do assay(sce, "exprs") or the equivalent.

Admittedly, this is all a bit confusing, so in the next release, I will remove the "exprs" output completely from readKallistoResults, i.e., only counts and TPMs are reported. If users want log-transformed TPMs, they can do it themselves and store the results in a "log_tpm" assay or something.

Personally, I don't think it makes much sense to add an offset of 1 when log-transforming CPM, TPM or FPKM values, see my rant here: https://support.bioconductor.org/p/104943/#104963

from scater.

Ah, I see now. Thanks for the detailed explanations!

from scater.