Phase 2 (calculate statistics)

#####Required Positional Arguments:

[1] Working Directory SelecT working directory created in Phase 1. Working Directory name can be changed but subdirectory names must be unchanged.

[2] Simulation Directory Directory where simulations can be found. Must contain simulation file neutral_simulation.tsv and selection_simulation.tsv. These can be found here: https://github.com/ridgelab/SelecT/tree/master/example/sim

[3] Chromosome Chromosome number where window can be found

[4] Window Number Window index number as defined by SelecT evironment setup. See SelecT_workspace/envi_files/all_wins for window ranges.

#####Optional Arguments: -inon Non-absolute iHS Runs iHS score probabilities where large negative scores ONLY are associated with selection (replicate CMS_local). Defualt is large positive AND negative iHS scores equate to greater selection (Voight).

--prior_prob Prior Probability Set Prior Probability to a custom value between 0.0 and 1. Defaults to (1 / actual number of variants within window).

--pp Prior Probability Flag Sets Prior Probability to 1/10,000 (replicate CMS_local).

--daf_cutoff DAF Cutoff Defines the derived dllele frequency cutoff for compose score. Defaults to a DAF value of 0.2. Special case: DAF value of 0.0 indicates incomplete MoP score calculation, PoP is unchanged.

Examples: java -Xmx[MB]m -jar StatsCalc.jar [1] [2] [3] [4] java -jar StatsCalc.jar example/SelecT_workspace example/sim 21 2

Note, this could be run by hand on a local machine, but we imagine automating the process a bit. A simple example SLURM script is provided (`selection.slurm') for running StatsCalc.jar for a single window. A possible bash script for submitting this SLURM script for each window to a high-performance computing cluster running SLURM might look like this:

#! /bin/bash

chromosome=21

for window in {0..3}
do
    sbatch selection.slurm $chromosome $window
done

exit 0