Exploiting Intrinsic Noise for Heterogeneous Cell Control Under Time Delays and Model Uncertainties

The majority of previous research in synthetic biology has focused on enabling robust control performance despite the presence of noise, while the understanding for how controllers may exploit that noise remains incomplete. Motivated by Maxwells Demon, we previously proposed a cellular control regime in which the exploitation of stochastic noise can break symmetry between and allow for specific control of multiple cells using a single input signal (i.e., single-input-multiple-output or SIMO control). The current work extends that analysis to include uncertain stochastic systems where system dynamics are are affected by time delays, intrinsic noises, and model uncertainty. We find that noise-exploiting controllers can remain highly effective despite coarse approximations to the models scale or incorrect estimations or extrinsic noise in key model parameters, and these controllers can even retain performance under substantial observer or actuator time delays. We also demonstrate how SIMO controllers could drive multi-cell systems to follow different trajectories with different phases and frequencies. Together, these findings suggest that noise-exploiting control should be possible even in the practical case where models are always approximate, where parameters are always uncertain, and where observations are corrupted by errors.