Michael Lawrence
Michael Lawrence
Here are two more ideas for annotations that are somewhat orthogonal to type checking and assertions: 1. Whether an argument is required or not. Some arguments without defaults can still...
It's common to want base-level (not run level) coverage from a GRanges object. It could be stored in a GPos, which behaves just like a GRanges, except there is a...
Current `read_bam()` returns a GAlignments or GAlignmentsList. I think we want to hide that data structure from the user. It could instead return a GRanges (GroupedGRanges for a GAlignmentsList) with...
Mapping tools like `mapToTranscripts()` from the GenomicFeatures package are very useful. Is it worth having wrappers `map_to_transcripts()` and `map_from_transcripts()` that use grouped GRanges instead of GRangesList?
Right now the user can get the transcript/exon structure from a BED file using the `blocks()` function, which converts the "blocks" column to a GRangesList. I think we'd rather have...
Microblocks based on blocks with that rely on multipass rendering do not render with the overlay. Would be nice if the microblocks would render in multiple passes in such cases.
It appears that on our cluster the port usage is too restrictive for the socket-based multicore backend. I think the base multicore uses raw connections or some sort of shared...
There are at least a couple inconsistencies with `torch.arange()`. In PyTorch: 1. Passing a `start` value without an `end` value generates the range `[0, start)`. That is convenient in the...
I would have assumed that `c()` would be equivalent to `rbind()` on SummarizedExperiment. However, it fails, because `bindROWS,Vector()` does not handle the assays. Perhaps `rbind,SummarizedExperiment()` should be refactored into a...
It looks like `helm-gtags-dwim` and `helm-gtags-find-tag-from-here` fail to find tags with a period (".") in their name. Perhaps this is due to how `thing-at-pt` parses symbols. Is there any way...