Added Chlamydomonas reinhardtii genome #863
Conversation
Codecov Report
@@ Coverage Diff @@
## main #863 +/- ##
==========================================
- Coverage 99.39% 99.38% -0.02%
==========================================
Files 74 58 -16
Lines 2498 2420 -78
Branches 285 284 -1
==========================================
- Hits 2483 2405 -78
Misses 7 7
Partials 8 8
Continue to review full report at Codecov.
|
|
Agreed we should do the selfing rate right - see #857. |
|
LGTM - can you squash the commits please @aays? https://stdpopsim.readthedocs.io/en/latest/development.html#rebasing |
|
Sounds good @jeromekelleher - on it! I've just heard back from Rob about per-chromosome mutation rates though, so I'm going to nab those as well before squashing the commits. Re: selfing rate, that sounds good as well - it's less so selfing and more infrequent sex altogether (Chlamy is happy to have many mitotic generations before periodically outcrossing) but I figure the same idea applies in that that's already baked into the estimate of Ne. |
|
Looks good! Merging. |
No demographic models for this one yet - I intend to do some digging after this, though, even if I'm not super confident there's many out there given how much Chlamy pop gen is only just taking off. But in either case, here's the reinhardtii genome!
An important note re: the recombination rate - sex takes place once every ~840 generations in the species (it's facultatively sexual), and the recombination rate estimates in here reflect that (e.g. they are r = rho / 2Ne, where rho was estimated directly from natural populations). To get 'true' physical recombination rate estimates, each of the values would be divided by the rate of sex, 0.001194 (explained further in the results/discussion here) which I've left as a note in the script. I figure that if simulating wild populations it makes more sense to have 'effective' recombination rates in there as opposed to the physical rates, since that would otherwise inflate the amount of recombination happening in the simulation.
If there's functionality to modify the rate of sex in a given simulation I think it'd make sense to encode the physical rates instead and save 0.001194 as a separate parameter. Happy to hear thoughts though!
Closes #828.