Luke Zappia

@duohongrui I have moved discussion about the BASiCS issue to #153

@azodichr Just checking where this is at?

@azodichr I have made the changes I suggested. Can you please double check I haven't broken anything before I merge this?

> To address your question... the Sys.sleep was an attempt at fixing the problem where when nCells.sample=TRUE it was sampling the nCells for each sample from the gamma distribution, but...

Hmmm...ok. I don't really have a use case other than trying to avoid errors. I do think it's pretty unexpected that something simple like this would fail though. I guess...

Yeah, I agree it could get complicated so up to maintainers whether it is worth the effort. The `datetime64` type came up as this is what **reticulate** converts R `datetime`...

This would mean transparency can't be used for edge weight. Might need to use size instead?

Yes, sorry if that wasn't clear. Because of how **{ggplot2}** works some things require the package to be loaded which affects **{clustree}**. See the linked issue for more details.

Allow direct passing of summarised node statistics for aesthetics, possible via a `node_metadata` argument. See #31 for example.

This approach doesn't work because you can't be sure what the aggregation function will do with `...`. For example `max()` uses it to accept all values to find the max...