Stochastic Computational Models for Gene Regulatory Networks and Dynamic Fault Tree Analysis

• Author / Creator

  Zhu, PeiCan

• Originally proposed in the 1960s, stochastic computation uses random binary bit streams to encode signal probabilities. Stochastic computation enables the implementation of basic arithmetic functions using simple logic elements. Here, the application of stochastic computation is extended to the domain of gene network models and the fault-tree analysis of system reliability.
  Initially, context-sensitive stochastic Boolean networks (CSSBNs) are developed to model the effect of context sensitivity in a genetic network. A CSSBN allows for a tunable tradeoff between accuracy and efficiency in a simulation. Studies of a simple p53-Mdm2 network reveal that random gene perturbation has a greater effect on the steady state distribution (SSD) compared to context switching activities. Secondly, stochastic multiple-valued networks (SMNs) are investigated to evaluate the effect of noise in a WNT5A network. Lastly, asynchronous stochastic Boolean networks (ASBNs) are proposed for investigating various asynchronous state updating strategies in a gene regulatory network (GRN). The dynamic behavior of a T helper network is investigated and the SSDs found by using ASBNs show the robustness of attractors of the network. In a long term, these results may help to accelerate drug discovery and develop effective drug intervention strategies for some genetic diseases.
  As another application of stochastic computation, the reliability analysis of dynamic fault trees (DFTs) is further pursued. Stochastic computational models are proposed for the priority AND (PAND) gate, the spare gate and probabilistic common cause failures (PCCFs). Subsequently, a phased-mission system (PMS) is analyzed by using a DFT to model each phase’s failure conditions. The accuracy of a stochastic analysis increases with the length of random binary bit streams in stochastic computation. In addition, non-exponential failure distributions and repeated events are readily handled by the stochastic computational approach. The accuracy, efficiency and scalability of the stochastic approach are demonstrated by several case studies of DFT analysis.

• Subjects / Keywords

  • Stochastic computing
  • Dynamic fault tree
  • Gene regulatory network

• Graduation date

  Fall 2015

• Type of Item

  Thesis

• Degree

  Doctor of Philosophy

• DOI

  https://doi.org/10.7939/R3M902B39

• License

  This thesis is made available by the University of Alberta Libraries with permission of the copyright owner solely for non-commercial purposes. This thesis, or any portion thereof, may not otherwise be copied or reproduced without the written consent of the copyright owner, except to the extent permitted by Canadian copyright law.