We propose a neural network based architecture to infer which parameters are fundamental, and their values, for producing specific instances of spatial patterns formed through cell colony growth. The system is trained on variations of the same pattern to recognize features that characterize it. Furthermore, selecting important parameters within our study mainly focuses on the fact that cells communicate. We use two forms of this communication as fundamental in finding the parameter values: bacterial conjugation, and environmental signals. The neural network is trained during 3000 epochs to identify the pattern class and specific parameter values needed to reproduce the desired pattern. These parameters are then inputted into a gro simulation to assess proximity to the original pattern. Our architecture achieved a 5% error upon pattern reproduction.

Synthetic Biology follows a Design-Build-Test-Learn cycle for the construction of novel biocircuits with a predefined behavior. Recently, a few simulation and bioCAD tools to assist synthetic biologists in the biocircuit engineering process have started to appear. One of these simulators is gro, a 2D Individual based Model (IbM). The objectives of our work on gro was to: Implement an IbM platform that can quickly simulate a large amount of E. Coli cells and to extend the array of functions that a cell executes. Our goal is to provide a fast and easy to use IbM simulator for synthetic circuits. We developed multiple extensions for gro as external modules programmed in C/C++. We improved the simulator speed through a new physics engine called CellEngine. It implements a ring-based algorithm for cell shoving. A “binary protein” module that directs intracellular regulation along with a new biocircuit language specification layer were added to the simulator as well. We also improved the existing signal capabilities in the simulator. Finally, bacterial conjugation (transmission of plasmids between neighbor bacteria) was implemented into gro as a new intercellular communication mechanism. Our new gro version enables the simulation of growing colonies having up to 105 bacteria in minutes; while the previous version took over five days to reach the same number. The new specification layer and the underlying regulatory module offer a new structure-oriented paradigm where cell behavior emerges from the specification. All of the new features extend gro to be a fast and versatile tool for prototyping multicellular biocircuits in microbial populations.