Hervé Pagès

Hello **ensembldb** people, FWIW a while ago I replaced `GenomeInfoDb::fetchExtendedChromInfoFromUCSC()` with `GenomeInfoDb::getChromInfoFromUCSC()`. The latter has the capability to map UCSC chromosome names to Ensembl names _as long as UCSC provides...

I'm not aware of an easy/reliable way to translate an Ensembl genome version to an UCSC genome. Could the data.frame returned by `registered_UCSC_genomes()` be used for that? E.g.: ``` library(GenomeInfoDb)...

Hi guys, I agree that unary `cbind()` and `rbind()` should be simplified. Just did that: https://github.com/Bioconductor/DelayedArray/commit/9198e3159e005d155ad176f0606ce051780bb7f4 `contentIsPristine()` means that the _content_ of the seeds (i.e. their array values) has not...

Also I'm open to suggestions for a better name for `contentIsPristine()`. `untouchedValues()`? `untouchedArrayValues()`? `valuesAreUntouched()`? `arrayValuesAreUntouched()`? `delayedOpsPreserveTheOriginalArrayValues()`? something else? Or should we have a function that does the opposite e.g. `valuesAreTouched()`?...

If we do something like this: ``` setMethod("%*%", c("DelayedMatrix", "DelayedMatrix"), function(x, y) { if (!(hasDelayedDimnames(x) || hasDelayedDimnames(y))) return(.BLOCK_matrix_mult(x, y)) x0

Yeah the "never touch the seed" principle in DelayedArray is a central one. No matter what operations are performed on `M`, the seeds of the result should always be identical...

> Is there any way to divert %*% from block-processing if the only delayed operations are subsetting or transposition? I hope so. In the meantime, I've added the "Implementing optimized...

Oh my! I hope you realize that changing the internal representation of DFrame objects will break all the serialized instances that exist around so will probably break hundreds of Bioconductor...

@jonocarroll Sounds good. If you move "grouping" to `mcols(x)` you don't need to touch `setClass` at all. @lawremi After adding a slot all the serialized instances become invalid. So it's...

> Good to hear that it's easy, because it was giving Gabe major headaches. Well the approach I'm proposing is simple and should be straightforward to implement. Still waiting for...