/Multicellular-Pattern-Synthesis

Primary LanguagePythonGNU General Public License v3.0GPL-3.0

Multicellular Pattern Synthesis

Container Contributors (alphabetical order)

Quicky Start Guide

This repository contains the code to replicate the computational environement for the publication by Demarcus Briers and Ashley Libby, Iman Haghighi, David Joy, Bruce Conklin, Calin Belta, and Todd McDevitt.

We provide a docker container to replicate our computaitonal environment on Windows, Mac, or Linux. Please see the docker folder for instructions on how install the required dependencies. Installing the dependencies using Docker is the quickest and most reproducible method of installing all software dependencies.

Once you have a working copy of the docker image (or have manually installed the dependencies), there are three types of analyis you can perform:

  • Simulations: For running the simulations with the computational model see the model folder.
  • Automated Pattern Design: For automating the deisgn of multicellular patterns with supervised learnings(TSSL) and mathematical optimization (Particle Swarm Optimization), see the synthesis folder.
  • Quantitative Pattern Verification: For quantifying patterns produced by simulations or experiements see the synthesis folder for a Machine Learning approach and the image_segmentation_clustering folder for a Computer Vision approach.