In emvolz-phylodynamics/mlesky: Maximum likelihood estimation of effective population size through time, growth rates, and phylodynamic regressions

Let's consider that the effective population was of size 10 except between 2000 and 2010 when it was of size 1 (bottleneck model). Ten samples per year are taken at uniform timepoints between 1990 and 2020. We can simulate such a phylogeny using:

library(ape)
library(mlesky)
set.seed(0)
alphaFun=function(x){if (x<2000 || x>2010) 10 else 1}
sampleDates=seq(1990,2020,0.1)
t=simCoal(sampleDates,alphaFun)
plotBoth(t,alphaFun)

We now infer the past population size given the phylogeny:

fit=mlskygrid(t,res=NULL,tau=NULL,tau_lower = 1,tau_upper = 10000,ncpu=2)
plot(fit)

Parametric bootstrap:

par <- parboot(fit)
plot(par)

emvolz-phylodynamics/mlesky documentation built on Feb. 13, 2025, 1:47 p.m.