Eugene-Oboh/Biofilm-Grazer-Model

Biofilm-Grazer Model is an agent based model I developed as part of my masters thesis in the marine biology MSc, at University of Bremen, Germany under the suoervision of Dr. Arjun Chennu, the head of Data Science &Technology working group at ZMT. The model predicts how nutrient(phosphorus), light and grazers affects the spatio-temporal development of riverine biofilms.

Biofilm Grazers Model

This repo tracks the model definitions and provides tools for running simulations using snakemake.

Installation

Use conda to create an environment with the dependencies in their latest versions

conda env create -f requirements.yml

To recreate exactly my environment:

conda env create -f environment.yml

Activate it with:

conda activate biofilm_grazers

The Snakefile points to a data folder in the same folder. So create a symlink to the data root folder:

ln -sf ../../path/to/data data  # use -f if you want to force change existing link

Running simulations

The major experiments for the simulations can be invoked with

snakemake -c1 run_experiments

Adapt the -c parameter to specify number of CPU cores to use.

This will run the model according to initial parameters specified in the model_param.json file Tests can be performed with a file at data/model_runs/trial_run/model_params.json, with contents:

{
  "run": {
    "simulation_minutes": 80640,
    "snapshot_interval": 30
  },
  "model": {
    "height": 20,
    "width": 20,
    "phosphorus_conc":35,
    "phosphorus_kp":3.5,
     "light_kl":0.1,
     "neighborhood_effect":false,
     
     "grazer_params": {
     "initial_gastropods":0,
      "grazing_rate_percent_of_biomass":2,
      "percent_of_food_to_weight":0.6,
      "init_grazer_biomass":3,
      "max_grazer_biomass":10
      },
    
    "biofilm_params": {
      "growth_rate": 0.0035,
      "max_biomass": 5,
      "initial_biomass_percent": [0,5]
    },
    "clock_params": {
      "step_size": 10
    }
  }
}

... or update the snakemake paths as fit for your file setup