Exploring Molecular Communication and Gene Regulation-Based Biocomputing of Bacteria

Sulakshana Somathilaka, Samitha (2024) Exploring Molecular Communication and Gene Regulation-Based Biocomputing of Bacteria. Doctoral thesis, SETU Waterford.

Text Samitha_SETU_PhD_Thesis__Revised_10152024.pdf Download (45MB)

Abstract

Artificial Intelligence (AI) has become a cornerstone of modern technological advancements, deeply intertwined with neuroscience and transforming into an essential part of daily life. AI has reshaped various industries, enhancing problem-solving capabilities and impacting societal norms. Originally inspired by the brain’s functions, such as neurons and synapses, AI has continually integrated neuroscience findings to improve systems’ sophistication and efficiency. This includes understanding brain plasticity and neuronal communication. Moreover, as AI has progressed, the focus has expanded from conventional neural networks to exploring neuromorphic architectures, including both silicon-based and biological systems, to enhance hardwarebased AI solutions. Although the integration of AI with silicon-based computing has significantly transformed society by enhancing efficiency and automating tasks across various sectors with minimal human input, this combination faces challenges such as high energy demands, complexity, adaptability, and biocompatibility. Therefore, this thesis explores the potential of bacteria as a living biocomputing platform. It begins with a macroscopic examination of bacterial communities’ collective behaviors and computational dynamics at the biome and population levels, which provides insights into their information processing, decision-making, and communication strategies. The focus then shifts to the single-cell level, specifically on the gene regulatory network (GRN) that drives bacterial computation. This investigation into the GRN reveals the cellular logic behind bacterial computing and paves the way for evaluating the reliability, energy efficiency, and practicality of bacterial systems for computational tasks like regression and classification. Highlighting the ultra-low energy dynamics of bacterial metabolism offers a solution to the energy limitations of silicon-based systems. Furthermore, the scalability, adaptability, and biocompatibility of bacterial populations address challenges in generalizing biological systems. The thesis aims to integrate these biological computing properties into conventional computing challenges, envisioning a transformative approach to AI and neuromorphic engineering through bacteria-based wet-neuromorphic systems, which could blend biological and computational intelligence.

Item Type:	Thesis (Doctoral)
Uncontrolled Keywords:	Molecular Communication, Gene Regulation-Based Biocomputing, Bacteria
Departments or Groups:	*NONE OF THESE*
Divisions:	School of Science > Department of Chemical and Life Sciences
Depositing User:	Derek Langford
Date Deposited:	21 Oct 2024 14:18
Last Modified:	21 Oct 2024 14:18
URI:	https://repository.wit.ie/id/eprint/7848

Actions (login required)

View Item