AGENT-BASED SIMULATION OF A COOPERATIVE MICROBIAL CONSORTIA REGULATED USING DUAL FEEDBACK CONTROL

Overview

GitHub repository for the final project of the Programmable Biology Course in the MSc in Computational Biology (2022-2023 course).

This project aims to model a a two-strain microbial consortia using the open-source software gro developed by the Klavins lab (Jang et al., 2012). This tool is an agent-based simulator that can predict cell-to-cell interactions between each member in a microbial consortium. Motivated by the results of previous studies demonstrating stable population compositions, we present a simulation of a two member community of bacteria that can maintain a steady-state population of both members via quorum sensing and a toxin-antitoxin system.

Dependencies

• gro beta.4
• R programming language

Running this project locally

1. Clone this project locally
2. Execute gro.exe file
3. Open one of the .gro simulation files
4. Run the simulation

Simulation files

There are 4 simulation files, one for each simulation performed for the project. The function used are the same, but some parameters such as the ccdB synthesis ratio or the initial number of cells vary from file to file to achieve different population behaviors:

• Stability: cooperation_data_stable.gro and cooperation_data.gro files.
• Different density ratio between the two members of the community: cooperation_data_differentRatio.gro file.
• Extinction of one strain: cooperation_data_extinction.gro

Simulated data

The data collected in the 4 simulations can be found in the Data folder contained in this respository. The results.R script can be used to generate plots representing the number of cells of each strain over the simulated time.

This work is licensed under a Creative Commons Attribution 3.0 Unported License.