RevBayes and TreePPL offer models to infer host repertoire evolution, but no tools to parse the outputs. evolnets has the necessary tools to reconstruct ancestral ecological networks based on posterior probabilities of interactions.
You can install evolnets like so:
if(!require("devtools", quietly = TRUE)) {
install.packages("devtools")
library(devtools)
} else {
library(devtools)
}
devtools::install_github("evonetslab/evolnets")The evolnets package provides three categories of important functions: rates, ancestral states and samples.
-
Rates: these functions are used to calculate effective rates of host-repertoire evolution:
effective_rate( ),count_events( ),rate_gl( ),count_gl( ). -
Ancestral states: these functions are used to calculate the posterior probabilities of host-parasite interactions at internal nodes of the parasite tree or at specific time points in the past:
posterior_at_nodes( ),posterior_at_ages( ). -
Samples: these functions perform calculations for each sampled host-parasite network during MCMC:
samples_at_ages( ),Q_posterior_at_ages( ),NODF_posterior_at_ages( ).
See the full documentation at the evolnets’ website.