The code requires the following python 3 packages: numpy, matplotlib, scipy

To view the simulation animation, python3 test.py.

To generate the plots used in our paper, run python3 generate_plots.py

1. Biomass vs longevity Over a range of p-values, for a given run, take the average biomass over every step, plotted against the longevity. Expect some maximum in the middle of biomass range.

   Q: How is this different from 3? Biomass(p_m)

2. GA - max p_m, convergence plot

   • Fitness functions
     1. f1(biomass) = average biomass / # cells
     2. f2(longevity) = longevity where longevity = the # of steps until all cells are empty, capped at 5000

3. Fitness landscape for p_m vs longevity, p_m vs biomass

   Evaluate over interval.

4. Introduce 2nd species

   • What p values do the different species evolve? We would expect p1, p2 = p / 2 p-value Distribution for single species, and for each of the 2 species (Show that they're different)
   • Covariance between the evolved growth rates of the 2 species

5a. # Fire fighters vs biomass, # fire fighters vs longevity

For p_m, with 1 or 2 species?

5b. Phase transitions in the # of firefighters in either longevity or biomass?

• Sharp transition, rate of change characterizes a phase transition, discontinuity

Plots 1,2,3 are with a single species Plots 1-4 without firefighters