simulation to quantify the prevalence of anomalous 4-taxon networks

how to run the pipeline

open file #0: edit as desired to set parameters.
run file #0 from command line, like this to run all first 3 steps:

Rscript 0_control_params.R

or like this to run step 1 only (network simulation with SiPhyNetwork). Replace 1 by 2 or by 3 to run step 2 only (network classification and gene tree simulation with PhyloCoalSimulations), or step 3 only (summarize all replicates from each scenario):

Rscript 0_control_params.R 1

check the number of networks simulated after step 1:

ls job[0-9]*/SiPhyNetwork_output/sim*.tree | wc -l # overall: should be #jobs * #nets
for jobdir in job[0-9]* # if more detail needed
do
  echo -n "$jobdir: "
  ls $jobdir/SiPhyNetwork_output/sim*.tree | wc -l
done

Step 4 files (starting with 4_*) are for sanity checks, and to compare the age of the simulated networks with the genetic distance threshold above which reticulation fails.

Step 5, to plot results (after sanity checks) and make figures using R: see 5_plot_results.Rmd.

Non-numbered .jl julia files: helper functions used in steps 2 and 4.

dependencies

SiPhyNetwork: now on CRAN, and other packages. Install within R with:

install.packages("SiPhyNetwork")
install.packages("tictoc")

PhyloNetworks and PhyloCoalSimulations: and others: all registered julia packages. They are listed, with specific version numbers, in Manifest.toml. Install them within Julia: first launch julia with julia --project to use Project.toml then install all packages with specific versions with this in package mode:

pkg> instantiate