Use this URL to cite or link to this record in EThOS:
Title: The application of nonlinear system identification in the field of synthetic biology
Author: Krishnanathan, Kirubhakaran
ISNI:       0000 0004 5358 2489
Awarding Body: University of Sheffield
Current Institution: University of Sheffield
Date of Award: 2014
Availability of Full Text:
Access from EThOS:
Access from Institution:
The field of synthetic biology has progressed from early concept, to initial demonstrations of simple genetic parts, and more recently to biological systems composed of functional modules that perform useful and specified tasks. Globally, there is an expectation that synthetic biology will deliver solutions to challenges, for instance, in healthcare, food security, and energy production. A key challenge in synthetic biology is to develop effective methodologies for characterisation of modular genetic parts in a form suitable for dynamic analysis and design. Dynamic analysis will enable the design of genetic parts to achieve robust and extensive functionalities, unlike the more commonly applied static analysis. In this thesis, improvements and new designs of both experimentation and modelling methods are presented, which were used for the quantitative analysis of transcriptional regulatory genetic parts and the development of mathematical models to aid predictive model-based design of higher-order genetic parts, in a top-down design approach. A data-driven nonlinear dynamic modelling framework is proposed to identify dynamic models of genetic parts. The identified models are shown to have compact representation and achieve rapid, accurate prediction of experimental data. The identification framework was extended by incorporating a computational Bayesian approach, to estimate the uncertainty of model parameters. The novel identification framework was used to capture the cell population heterogeneity observed in experimental data of the systems. To investigate if a reporter cascade has an influential effect on the dynamics of the system to which it has been linked to, the identification framework was used to characterise dynamics of two transcriptional regulatory systems - the same functional module but different reporter cascades. For the first time this provided evidence that the reporter cascades do have an influential effect on the dynamics of the systems. Generalised frequency response functions obtained from the identified dynamic models provided an alternative tool for dynamic characterisation of genetic parts which could be used for design purposes. In addition, characterising only the functional module - BBa_F2620 relative to a reporter cascade was found to be unachievable using the implemented experimentation. However, with the identification and analysis tools used, the commonality of the systems under investigation is retrieved and adequately characterised.
Supervisor: Kadirkamanathan, Visakan ; Billings, Stephen Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID:  DOI: Not available