Windowed mutation rate over the genome?

[hyanwong]

I'm having a senior moment here: how can I calculate a windowed mutation rate along the genome? I mean, if I want it per tree, I could do:

rates = []
for tree in ts.trees():
    rates.append(tree.num_mutations / tree.total_branch_length)

or if I want it across the whole genome, I can do:

av_rate = ts.num_mutations / np.sum(
    (ts.edges_right - ts.edges_left) * (ts.nodes_time[ts.edges_parent] - ts.nodes_time[ts.edges_child])
)

But what's the magic stats incantation to do this in e.g. 100kb windows?

[nspope]

i'd just divide windowed site-mode segregating sites by windowed branch-mode segregating sites

[nspope]

though, that's assuming biallelic sites-- would take a bit more thought with multiallelic sites.

[hyanwong]

Ah yes, this would need the "mutation mode" thingy. Thanks

[hyanwong]

I think some people might find a windowed mutation rate plot quite useful. We should describe this in the stats docs.