Toggle Main Menu Toggle Search

Open Access padlockePrints

Modelling Bacterial Regulatory Networks with Petri Nets

Lookup NU author(s): Oliver Shaw



To exploit the vast data obtained from high throughput molecular biology, a variety of modelling and analysis techniques must be fully utilised. In this thesis, Petri nets are investigated within the context of computational systems biology, with the specific focus of facilitating the creation and analysis of models of biological pathways. The analysis of qualitative models of genetic networks using safe Petri net techniques was investigated with particular reference to model checking. To exploit existing model repositories a mapping was presented for the automatic translation of models encoded in the Systems Biology Markup Language (SBML) into the Petri Net framework. The mapping is demonstrated via the conversion and invariant analysis of two published models of the glycolysis pathway. Dynamic stochastic simulations of biological systems suffer from two problems: computational cost; and lack of kinetic parameters. A new stochastic Petri net simulation tool, NASTY was developed which addresses the prohibitive real-time computational costs of simulations by using distributed job scheduling. In order to manage and maximise the usefulness of simulation results a new data standard, TSML was presented. The computational power of NASTY provided the basis for the development of a genetic algorithm for the automatic parameterisation of stochastic models. This parameter estimation technique was evaluated on a published model of the general stress response of E. coli. An attempt to enhance the parameter estimation process using sensitivity analysis was then investigated. To explore the scope and limits of applying the Petri net techniques presented, a realistic case study investigated how the Pho and sB regulons interact to mitigate phosphate stress in Bacillus subtilis. This study made use of a combination of qualitative and quantitative Petri net techniques and was able to confirm an existing experimental hypothesis. This work was supervised by Anil Wipat and Jason Steggles.

Publication metadata

Author(s): Shaw OJ

Publication type: Report

Publication status: Published

Series Title: School of Computing Science Technical Report Series

Year: 2007

Pages: 278

Print publication date: 01/09/2007

Source Publication Date: September 2007

Report Number: 1050

Institution: School of Computing Science, University of Newcastle upon Tyne

Place Published: Newcastle upon Tyne