Stochastic analysis of synthetic genetic circuits

Update item information
Publication Type dissertation
School or College College of Engineering
Department Computing (School of)
Author Madsen, Curtis K.
Title Stochastic analysis of synthetic genetic circuits
Date 2013-08
Description Over the past few decades, synthetic biology has generated great interest to biologists and engineers alike. Synthetic biology combines the research of biology with the engineering principles of standards, abstraction, and automated construction with the ultimate goal of being able to design and build useful biological systems. To realize this goal, researchers are actively working on better ways to model and analyze synthetic genetic circuits, groupings of genes that influence the expression of each other through the use of proteins. When designing and analyzing genetic circuits, researchers are often interested in building circuits that exhibit a particular behavior. Usually, this involves simulating their models to produce some time series data and analyzing this data to discern whether or not the circuit behaves appropriately. This method becomes less attractive as circuits grow in complexity because it becomes very time consuming to generate a sufficient amount of runs for analysis. In addition, trying to select representative runs out of a large data set is tedious and error-prone thereby motivating methods of automating this analysis. This has led to the need for design space exploration techniques that allow synthetic biologists to easily explore the effect of varying parameters and efficiently consider alternative designs of their systems. This dissertation attempts to address this need by proposing new analysis and verification techniques for synthetic genetic circuits. In particular, it applies formal methods such as model checking techniques to models of genetic circuits in order to ensure that they behave correctly and are as robust as possible for a variety of different inputs and/or parameter settings. However, model checking stochastic systems is not as simple as model checking deterministic systems where it is always known what the next state of the system will be at any given step. Stochastic systems can exhibit a variety of different behaviors that are chosen randomly with different probabilities at each time step. Therefore, model checking a stochastic system involves calculating the probability that the system will exhibit a desired behavior. Although it is often more difficult to work with the probabilities that stochastic systems introduce, stochastic systems and the models that represent them are becoming commonplace in many disciplines including electronic circuit design where as parts are being made smaller and smaller, they are becoming less reliable. In addition to stochastic model checking, this dissertation proposes a new incremental stochastic simulation algorithm (iSSA) based on Gillespie's stochastic simulation algorithm (SSA) that is capable of presenting a researcher with a simulation trace of the typical behavior of the system. Before the development of this algorithm, discerning this information was extremely error-prone as it involved performing many simulations and attempting to wade through the massive amounts of data. This algorithm greatly aids researchers in designing genetic circuits as it efficiently shows the researcher the most likely behavior of the circuit. Both the iSSA and stochastic model checking can be used in concert to give a researcher the likelihood that the system will exhibit its most typical behavior. Once the typical behavior is known, properties for nontypical behaviors can be constructed and their likelihoods can also be computed. This methodology is applied to several genetic circuits leading to new understanding of the effects of various parameters on the behavior of these circuits.
Type Text
Publisher University of Utah
Subject Design space exploration; Stochastic model checking; Stochastic simulation; Synthetic biology
Dissertation Institution University of Utah
Dissertation Name Doctor of Philosophy
Language eng
Rights Management Copyright © Curtis K. Madsen 2013
Format Medium application/pdf
Format Extent 3,524,701 bytes
Identifier etd3/id/2537
ARK ark:/87278/s6vf06m9
Setname ir_etd
Date Created 2013-11-22
Date Modified 2017-10-26
ID 196113
Reference URL https://collections.lib.utah.edu/ark:/87278/s6vf06m9