2. I don’t quite get this. My theoretical work computes expected psi for a single snp/mutation, so I don’t see how molecular clock (based on accumulation of mutations?) work here. I’ve linked the preprint in another post which gives more details.

1. For coalescent computations it is enough to consider 2 samples from A and 2 samples from B, which is what I did. With data, really you can use as much as you want to compute frequencies.

anyway, try it out and let me know what you think! i would appreciate any feedback and/or bug reports uwu

4/4

the main caveat is that slim-runner can only currently deal with tabular output in the csv format. it will work great if you are recording a set of quantities per cycle or per simulation. but, it won't help you directly if you are trying to e.g. keep track of tree sequences

3/

this wrapper is inspired by @vsbuffalo.bsky.social's slimflow, but has two main differences:

a) it's not based on snakemake and therefore can easily handle 10s or 100s of millions of simulations, and

b) it batches output to disk so you don't end up with millions of tiny files

2/