Access Type

Open Access Dissertation

Date of Award

January 2018

Degree Type

Dissertation

Degree Name

Ph.D.

Department

Electrical and Computer Engineering

First Advisor

Harpreet Singh

Abstract

DNA is the basic building block of any living organism. DNA is considered a popular candidate for future biological devices and circuits for solving genetic disorders and several other medical problems. With this objective in mind, this research aims at developing novel approaches for the design of DNA based circuits. There are many recent developments in the medical field such as the development of biological nanorobots, SMART drugs, and CRISPR-Cas9 technologies. There is a strong need for circuits that can work with these technologies and devices. DNA is considered a suitable candidate for designing such circuits because of the programmability of the DNA strands, small size, lightweight, known thermodynamics, higher parallelism, and exponentially reducing the cost of synthesizing techniques. The DNA strand displacement operation is useful in developing circuits with DNA strands. The circuit can be either a digital circuit, in which the logic high and logic low states of the DNA strand concentrations are considered as the signal, or it can be an analog circuit in which the concentration of the DNA strands itself will act as the signal.

We developed novel approaches in this research for the design of digital, as well as analog circuits keeping in view of the number of DNA strands required for the circuit design. Towards this goal in the digital domain, we developed spatially localized DNA majority logic gates and an inverter logic gate that can be used with the existing seesaw based logic gates. The majority logic gates proposed in this research can considerably reduce the number of strands required in the design. The introduction of the logic inverter operation can translate the dual rail circuit architecture into a monorail architecture for the seesaw based logic circuits. It can also reduce the number of unique strands required for the design into approximately half. The reduction in the number of unique strands will consequently reduce the leakage reactions, circuit complexity, and cost associated with the DNA circuits.

The real world biological inputs are analog in nature. If we can use those analog signals directly in the circuits, it can considerably reduce the resources required. Even though analog circuits are highly prone to noise, they are a perfect candidate for performing computations in the resource-limited environments, such as inside the cell. In the analog domain, we are developing a novel fuzzy inference engine using analog circuits such as the minimum gate, maximum gate, and fan-out gates. All the circuits discussed in this research were designed and tested in the Visual DSD software. The biological inputs are inherently fuzzy in nature, hence a fuzzy based system can play a vital role in future decision-making circuits. We hope that our research will be the first step towards realizing these larger goals. The ultimate aim of our research is to develop novel approaches for the design of circuits which can be used with the future biological devices to tackle many medical problems such as genetic disorders.

Download

Share

COinS