Supporting code for the article:

J Yu and N Bagheri. (2020). Agent-based models predict emergent behavior of heterogeneous cell populations in dynamic microenvironments. Frontiers in Bioengineering and Biotechnology. doi: 10.3389/fbioe.2020.00249

Setup files

The setups directory contains all the setup files used for running simulations. Simulations were run using ARCADE v2.2.

The simulations DEFAULT_random, MODULE_COMPLEXITY_both, MODULE_COMPLEXITY_metabolism, and MODULE_COMPLEXITY_signaling use modified code; see supplementary materials for details.

Simulation data

Raw simulation data and results are available on Mendeley Data:

• DEFAULT . http://dx.doi.org/10.17632/wzs7pxkgb9
• MODULE COMPLEXITY . http://dx.doi.org/10.17632/7w2cdsrt87
• PARAMETER SENSITIVITY . http://dx.doi.org/10.17632/wpfycmb5sv
• GROWTH CONTEXT . http://dx.doi.org/10.17632/mmnh9hsv5y
• CELL COMPETITION (1/3) . http://dx.doi.org/10.17632/5h6ng2y4fc
• CELL COMPETITION (2/3) . http://dx.doi.org/10.17632/gzfwhgdtwz
• CELL COMPETITION (3/3) . http://dx.doi.org/10.17632/dbj46mw6j9
• POPULATION HETEROGENEITY (1/5) . http://dx.doi.org/10.17632/bk9s769t3t
• POPULATION HETEROGENEITY (2/5) . http://dx.doi.org/10.17632/mjpcm5my8z
• POPULATION HETEROGENEITY (3/5) . http://dx.doi.org/10.17632/hskf9mxbpw
• POPULATION HETEROGENEITY (4/5) . http://dx.doi.org/10.17632/ynzsfdswjz
• POPULATION HETEROGENEITY (5/5) . http://dx.doi.org/10.17632/p46dwfyvzg

Pipeline notebooks

Parse simulation outputs

The parse_simulation_outputs notebook provides the functions and scripts for parsing simulation files (.json) into pickled numpy arrays (.pkl). These parsed results are included with the raw simulation data.

Analyze data & results

The analyze_data_results notebook provides functions and scripts for running basic analysis on simulation data and parsed results. All resulting .json and .csv files are provided in the analysis directory.

Generate figure inputs

The generate_figure_inputs notebook walks through all the steps necessary to generate figure input files from raw data, parsed files, and basic analysis files. All resulting files are provided in the analysis directory. Refer to figure section in notebook for more details.

To view figures, start a local HTTP server from the root folder, which can be done using Python or PHP:

$ python3 -m http.server
$ php -S 127.0.0.1:8000

Note that the links in the notebook to figures assume the local port 8000; if your server is running on a different port, the links to the figures from the notebook will not work. Instead, you can navigate to http://localhost:XXXX/ where XXXX is the port number and follow links to the figures.