Entrainment and Control of Bacterial Populations: An in Silico Study over a Spatially Extended Agent Based Model
ACS Synthetic Biology
American Chemical Society
Reason for embargo
We extend a spatially explicit agent based model (ABM) developed previously to investigate entrainment and control of the emergent behavior of a population of synchronized oscillating cells in a microfluidic chamber. Unlike most of the work in models of control of cellular systems which focus on temporal changes, we model individual cells with spatial dependencies which may contribute to certain behavioral responses. We use the model to investigate the response of both open loop and closed loop strategies, such as proportional control (P-control), proportional-integral control (PI-control) and proportional-integral-derivative control (PID-control), to heterogeinities and growth in the cell population, variations of the control parameters and spatial effects such as diffusion in the spatially explicit setting of a microfluidic chamber setup. We show that, as expected from the theory of phase locking in dynamical systems, open loop control can only entrain the cell population in a subset of forcing periods, with a wide variety of dynamical behaviors obtained outside these regions of entrainment. Closed-loop control is shown instead to guarantee entrainment in a much wider region of control parameter space although presenting limitations when the population size increases over a certain threshold. In silico tracking experiments are also performed to validate the ability of classical control approaches to achieve other reference behaviors such as a desired constant output or a linearly varying one. All simulations are carried out in BSim, an advanced agent-based simulator of microbial population which is here extended ad hoc to include the effects of control strategies acting onto the population.
The authors declare no competing interests. We thank Dr. Nigel J. Savery at the University of Bristol for useful discussions around the subject of GRNs and for his help in developing the original ABM model. We also wish to thank Dr Gianfranco Fiore at the University of Bristol and the anonymous reviewers for reading the revised manuscript carefully and providing insightful comments that led to a consistent revision of the original manuscript. P.M. was supported by EPSRC Grant EP/E501214/1 and K.T.-A. by EPSRC Grant EP/I018638/1. The funders had no role in study design, data collection, and analysis, decision to publish, or preparation of the manuscript. This work was carried out using the computational facilities of the Advanced Computing Research Centre, University of Bristol, http://www.bris.ac.uk/acrc/.
This is the author accepted manuscript. The final version is available from American Chemical Society via the DOI in this record.