|M.Sc Student||Abo Hanna Hanna|
|Subject||Nano-Electronics for Modeling Synthetic and System Biology|
|Department||Department of Biomedical Engineering||Supervisor||ASSOCIATE PROF. Ramez Daniel|
|Full Thesis text|
Cytomorphic engineering attempts to study the cellular behavior of biological systems using electronics. As such, it can be considered analogous to the study of neurobiological concepts for neuromorphic engineering applications. The main goal of cytomorphic engineering is to simulate cells, organs, and tissues while considering the stochastic behavior of a single cell, cell-to-cell variation, distortion, and crosstalk using mixed-signal integrated electronics. Such simulations are computationally intensive and can take weeks using modern digital hardware. Additionally, cytomorphic electronics are used to design novel large-scale synthetic biological systems by providing a fast and simple emulative framework. Cytomorphic circuits can also map architectural concepts and design principles from cells to electronics. Living cells have the ability to perform complex, real-time and highly sensitive tasks and process environmental input signals with highly noisy and imprecise parts, such that reliable outputs are produced. These properties make them the ultimate candidate for designing noise-tolerant, ultra-low power electronic systems.
To date, digital and analog translinear electronics have commonly been used in the design of cytomorphic circuits; Such circuits could greatly benefit from lowering the area of the digital memory via memristive circuits. In this thesis, we propose an alternative approach that utilizes the emergent properties of nano-electronics. We show that two-terminal memristive devices can capture the non-linear and stochastic behavior of biochemical reactions. Memristive devices can retain a state of internal resistance based on the history of the applied voltage and the current flowing through them. In the last decade, memristive devices have been proposed in a broad range of applications, including but not limited to resistive random-access memory, neuromorphic systems, Boolean logic gates and programmable analog circuits. The analogy of memristive devices and biochemical binding reactions is made at the biophysical dynamic and energy level. Both have an input-output transfer function with a non-linear behavior and controlled by time-dependent internal state variables. The two logic states of digital memristor can represent the binding and unbinding reactions of gene networks. The dynamics of the enzymatic reaction and forming a new complex, and the dynamics of switching a memristor all follow the Poisson distribution.
In the thesis, we present the design of several building blocks based on analog memristive circuits that inherently model the biophysical mechanisms of gene expression. The circuits model induction by small molecules, activation and repression by transcription factors, biological promoters, cooperative binding, and transcriptional and translational regulation of gene expression. Finally, we utilize the building blocks to form complex mixed-signal networks that can simulate the delay-induced oscillator and the p53-mdm2 interaction in the cancer signaling pathway. By capitalizing on the common themes and emerging bridges between biochemical reactions and memristive devices, the “cytomorphic” mapping between cellular biology and memristors leads to the design of a fast and simple emulative framework for studying genetic circuits and arbitrary large-scale biological networks in systems and synthetic biology.