Repository navigation
Comparison to "Threads" #519
hyanwong
started this conversation in
Show and tell
Replies: 2 comments
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment


Uh oh!
There was an error while loading. Please reload this page.
Uh oh!
There was an error while loading. Please reload this page.
I thought it would be interesting to compare how tsinfer+tsdate do against the new "Threads" program (https://pypi.org/project/threads-arg/0.1.0/).
As a test, I have a stdpopsim model with 2 chimp species (3 populations) and some selective sweeps in 2 of the populations:
Here are the edge plots from the original vs tsinfer+tsdate and Threads on 5Mb of genome: the 2 recent selective sweeps (in western at 1/3rd along the genome, and bonobo at 2/3rds) are obvious in the plot of the true edges. Threads seems a little better in picking up the recent sweeps, but maybe picks up less of the demographic banding at the top (although to be fair, I told it there was a fixed population size of 100,000 haploid genomes). For this scale of data (120 genomes), Threads is slightly faster than tsinfer (and it is relatively faster if the genome length increases, but I suspect we would see a relative slow down if the sample size increased to very large numbers)
Click for code
All reactions