tskit-dev / pyslim

Tools for dealing with tree sequences coming to and from SLiM.

MIT License

27 stars 23 forks source link

WIP Union vignette #190

Closed mufernando closed 3 years ago

mufernando commented 3 years ago

WIP

petrelharp commented 3 years ago

I made some minor tweaks - but, I see it's not quite done at the end? Looking for the elegant way to do that last recursive union?

mufernando commented 3 years ago

I made some minor tweaks - but, I see it's not quite done at the end? Looking for the elegant way to do that last recursive union?

Thanks! I did the union recursion starting here. We could do some sort of analysis at the end to show it worked, is that what you mean?

codecov-commenter commented 3 years ago

Codecov Report

Merging #190 (dbdc29c) into main (a1d5588) will not change coverage. The diff coverage is n/a.

@@           Coverage Diff           @@
##             main     #190   +/-   ##
=======================================
  Coverage   87.84%   87.84%           
=======================================
  Files           7        7           
  Lines         987      987           
  Branches      188      188           
=======================================
  Hits          867      867           
  Misses         88       88           
  Partials       32       32

Continue to review full report at Codecov.

Legend - Click here to learn more Δ = absolute <relative> (impact), ø = not affected, ? = missing data Powered by Codecov. Last update a1d5588...dbdc29c. Read the comment docs.

mufernando commented 3 years ago

I think this is ready to be reviewed. I included two bits at the end that make use of the union-ed tree sequence:

Count the number of samples in each of the independently simulated simulations and make sure that checks out
Added a figure with some trees illustrating ILS (I made one of the branches shorter and with a lot of individuals

Still think we might need to work on the text a bit, maybe to make it more detailed?

petrelharp commented 3 years ago

Ah, one more thing - I don't think we should be checking in the .trees files (although I notice that previous ones are, hm).

petrelharp commented 3 years ago

... well, I guess we have the .trees files in for good reason. (CI breaks otherwise)

mufernando commented 3 years ago

I think CI can't run make -- or at least not with more than one core.

petrelharp commented 3 years ago

I think CI can't run make -- or at least not with more than one core.

Ah, would that actually make the file not available when it's supposed to be? The test passes otherwise.