Use this URL to cite or link to this record in EThOS:
Title: Moment closure and parameter estimation in stochastic biological models
Author: Milner , Peter
ISNI:       0000 0004 0125 2074
Awarding Body: University of Newcastle Upon Tyne
Current Institution: University of Newcastle upon Tyne
Date of Award: 2011
Availability of Full Text:
Access from EThOS:
One of the key problems in the new science of systems biology is inferring rate parameters for complex stochastic kinetic biochemical network models, using partial and discrete time- course measurements of the system state. Although exact inference for stochastic models is possible (Boys et al., 2008), it is computationally intensive for relatively small networks. We explore the Bayesian estimation of stochastic kinetic rate parameters using approximate methods, based on moment closure analysis of the underlying stochastic process. Moment closure has been widely studied and applied in many areas. In addition to using previous methodology, we introduce a technique to perform moment closure on models with rational rate laws, allowing applications to a larger class of models. By assuming a Gaussian distribution and using moment-closure estimates of the first two moments, we can greatly increase the speed of parameter inference. The parameter space can be efficiently explored by embedding this approximation into an MCMC procedure. Mixing problems often occur when estimating latent data values. We tackle this problem using a block updating approach conditioning on all adjacent data points. Sporulation in the bacteria Bacillus subtilis is a well studied mechanism of survival, yet in practice there are large costs associated with gene knockouts and laboratory experiments. We built a stochastic model of this process allowing for external factors that trigger sporulation, such as the response from the gene lexA to external factors endangering Bacillus subtilis. The calibration of the model is done with fluorescence microscopy data readings of levels of the GFP protein (Veening et al., 2009). Predictions were made from the model as to which gene knockouts would have large effects on the proportion of Bacillus subtilis that would sporulate.
Supervisor: Not available Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID:  DOI: Not available