Regulation strategies for two-output biomolecular networks

Abstract

Feedback control theory facilitates the development of self-regulating systems with desired performance which are predictable and insensitive to disturbances. Feedback regulatory topologies are found in many natural systems and have been of key importance in the design of reliable synthetic bio-devices operating in complex biological environments. Here, we study control schemes for biomolecular processes with two outputs of interest, expanding previously described concepts based on single-output systems. Regulation of such processes may unlock new design possibilities but can be challenging due to coupling interactions; also potential disturbances applied on one of the outputs may affect both. We therefore propose architectures for robustly manipulating the ratio/product and linear combinations of the outputs as well as each of the outputs independently. To demonstrate their characteristics, we apply these architectures to a simple process of two mutually activated biomolecular species. We also highlight the potential for experimental implementation by exploring synthetic realizations both in vivo and in vitro . This work presents an important step forward in building bio-devices capable of sophisticated functions.

AC-BioSD : A biomolecular signal differentiator module with enhanced performance (extended version)

“…A promising candidate might be a Fig. 1: (a) Biochemical reactions considered in this work [26]. (b) BioSD-III module with U and X 1 being the input and the output signal/species, respectively [19].…”

“…We use three main types of biochemical reactions to structurally describe the biomolecular topologies herein, which Fig. 1: (a) Biochemical reactions considered in this work [25]. (b) BioSD-III module with U and X 1 being the input and the output signal/species, respectively [19].…”

“…4(a) illustrates how derivative control can be realized via the AC-BioSD module within the PID control scheme presented in Section II. C. For demonstration purposes, we consider a simple biological process of two mutually activated species [17], [26] inside the cloud network (this does not affect the generality of the PID architecture outside the cloud network). The corresponding closed-loop dynamics is given by the ODE model:…”

“…C. For demonstration purposes, we consider a simple biological process of two mutually activated species [17], [26] inside the cloud network (this does not affect the generality of the PID architecture outside the cloud network). The corresponding closed-loop dynamics is given by the ODE model: where F = γ 1 − γ 3 Y + γ 6 W and .…”

“…We use three main types of biochemical reactions to structurally describe the biomolecular topologies herein, which are aligned with the law of mass action [19], [25], [26]: general transformation of reactants into products, catalytic production and catalytic inhibition. Assuming A, B are two bio-chemical species, these reactions take the following general form A → B, A → A + B, A + B → A , respectively.…”