tskit-dev / msprime-1.0-paper

Publication describing msprime 1.0
4 stars 20 forks source link

Contrast sim_mutations performance with existing methods #203

Closed jeromekelleher closed 2 years ago

jeromekelleher commented 3 years ago

Reviewer 2 says:

As another example, line 301-4 states that "Simulating mutations ... is efficient" and refers to Fig 3 for evidence. The rest of the section (304-319) details a number of examples shown in Fig 3 that do not add much to the story.

We originally avoided contrasting the performance of sim_mutations with Pyvolve and SeqGen because it's hard to disentangle the method performance from inefficient file formats. The concrete examples were intended to sidestep this by showing that sim_ancestry is as fast as you'd ever need it to be, but this point was lost I think. So, better to use the space to run comparable mutation simulations in Pyvolve and SeqGen and report how long they took.