Description

Forward engineering of synthetic genetic circuits-based artificial information

processing devices, termed synthetic genetic devices, has been carried out by applying

hierarchical computer and electronic design principles to biology in order to

accomplish cellular biocomputing. Such devices with high computing power can

process higher level information in microbes and mammalian cells. Expansion of the

computing power of these devices with increasing complexity might lead to the

solving of complex computing problems in certain areas where living cell-based

biocomputing might outperform traditional computers.

This thesis centrally focuses on distributed computing in engineered bacteria. Here

artificial neural network (ANN) is adapted as a computing system in living

Escherichia coli cells to develop a design framework for building complex computing

functions including a 2-to-4 decoder and a 4-to-2 priority encoder. The basic concept

of single layer ANN architecture is mapped into engineered E. coli cells where

individual engineered bacterial cells, termed bactoneurons, carry cellular devices and

act as artificial neuro-synapses. A complete set of rules is established to distribute the

truth table of a function among multiple fragmented truth tables followed by the

derivation of unit bactoneurons from those fragmented truth tables without

considering the electronic circuit design of that function. When those bactoneurons are

cultured together, they work similar to a single layer ANN type architecture and give

rise to the original computing function.