Closed petrelharp closed 5 years ago
I'm good with this --- I think the paper is strong even without the real data, and it's perfectly reasonable to not include stuff from tsinfer, since it doesn't do branch lengths.
However, we could keep the plot of branch vs site stats just for Relate. This isn't much work, and it is an interesting thing, right?
LGTM. i'll do another pass-thru after a merge?
This is superceded by #24 .
Here's the paper without the 1000 genomes data. 17 pages.