서브메뉴
검색
Analytical and Computational Approaches to Nonlinear Dynamics in Enzyme Kinetics and Sleep-Wake Regulation
Analytical and Computational Approaches to Nonlinear Dynamics in Enzyme Kinetics and Sleep-Wake Regulation
상세정보
- 자료유형
- 학위논문 서양
- 최종처리일시
- 20260202105224
- ISBN
- 9798291566428
- DDC
- 574
- 서명/저자
- Analytical and Computational Approaches to Nonlinear Dynamics in Enzyme Kinetics and Sleep-Wake Regulation
- 발행사항
- [Sl] : University of Michigan, 2025
- 발행사항
- Ann Arbor : ProQuest Dissertations & Theses, 2025
- 형태사항
- 240 p
- 주기사항
- Source: Dissertations Abstracts International, Volume: 87-02, Section: B.
- 주기사항
- Advisor: Booth, Victoria;Schnell, Santiago.
- 학위논문주기
- Thesis (Ph.D.)--University of Michigan, 2025.
- 초록/해제
- 요약Reduced models for enzymatic mechanisms, widely used in real-world biochemical applications, are often governed by parametric restrictions that determine their accuracy and applicability. The work in this thesis focuses on addressing challenges in analyzing solutions of reduced enzyme kinetic models, particularly in settings where assumptions of traditional analytic methods may fail. A second focus of this thesis is the development of a gradient-based optimization framework for fitting noisy differential equation models for the neuronal regulation of sleep-wake behavior to biological data.In Chapter 2, we derive slow-scale linear noise approximation (ssLNA) to stochastic enzyme kinetic models directly using geometric singular perturbation theory (GSPT). We resolve discrepancies among prior stochastic reductions by identifying the role of the Tikhonov standard form in producing consistent reductions. Extending the ssLNA to systems with a transcritical bifurcation, we disprove the widely-used Segel-Slemrod condition for stochastic model reduction in the Michaelis-Menten (MM) mechanism. Instead, we demonstrate that the more restrictive Reich-Sel'kov condition is needed, aligning stochastic and deterministic validity thresholds for quasi-steady-state approximations (QSSAs).In Chapter 3, we focus on the MM model to investigate the parametric conditions under which the standard QSSA is not only valid but also predominant among competing reductions. Using the theory of fences and anti-funnels, we locate the slow manifold in the MM phase plane and derive sharp error bounds. While the Reich-Sel'kov condition governs accuracy, we introduce a new and more restrictive qualifier that predicts the predominance of the standard QSSA. This distinction is critical when multiple reduced models are valid, as it guides optimal model selection for experimental design.In Chapter 4, we perform model reduction analysis for the Intermolecular Autocatalytic Zymogen Activation (IAZA) model, when it features a dynamic transcritical bifurcation. GSPT fails in this setting due to loss of normal hyperbolicity. Using the blow-up method, we transform the system near the bifurcation to recover hyperbolicity and derive an explicit solution to the desingularized system at the bifurcation point. This allows us to estimate the peak complex concentration analytically and explain the surprising accuracy of the nullcline-based reduction. We also identify the existence of a canard and analyze the implications of dynamic imperfections for model reduction.Chapter 5 tackles perturbation systems that lack an explicit Tikhonov standard form structure but mimic typical standard form behavior portraying fast convergence to QSS approximations. We propose a novel use of the blow-up method to connect the fast and slow dynamics in systems where a clear distinction is not apparent. Applied to the MM and IAZA systems, this approach reveals that the systems admit latent standard fast-slow structure and the trajectories starting both on and away from the critical manifold get attracted to the QSS curves, validating the resemblance to standard form dynamics.Chapter 6 shifts focus to computational neuroscience and presents a framework for fitting noisy differential equation models to rodent sleep-wake data using gradient-based optimization. We adapt a mechanistic model of interacting neuronal populations with noisy firing activity. Leveraging automatic differentiation, maximum mean discrepancy, and sensitivity analysis, we demonstrate efficient parameter inference from discrete sleep score data. The methodology offers a framework to relate experimental hypotheses for the neuronal circuit structures responsible for sleep-wake regulation to the random polyphasic sleep-wake behavior of rodents, with applications in and beyond sleep modeling.
- 일반주제명
- Biology
- 일반주제명
- Mathematics
- 일반주제명
- Applied mathematics
- 일반주제명
- Biochemistry
- 기타저자
- University of Michigan Applied and Interdisciplinary Mathematics
- 기본자료저록
- Dissertations Abstracts International. 87-02B.
- 전자적 위치 및 접속
- 로그인 후 원문을 볼 수 있습니다.
MARC
008260126s2025 us c eng d■001000017359846
■00520260202105224
■006m o d
■007cr#unu||||||||
■020 ▼a9798291566428
■035 ▼a(MiAaPQ)AAI32271830
■035 ▼a(MiAaPQ)umichrackham006360
■040 ▼aMiAaPQ▼cMiAaPQ
■0820 ▼a574
■1001 ▼aSrivastava, Kashvi.
■24510▼aAnalytical and Computational Approaches to Nonlinear Dynamics in Enzyme Kinetics and Sleep-Wake Regulation
■260 ▼a[Sl]▼bUniversity of Michigan▼c2025
■260 1▼aAnn Arbor▼bProQuest Dissertations & Theses▼c2025
■300 ▼a240 p
■500 ▼aSource: Dissertations Abstracts International, Volume: 87-02, Section: B.
■500 ▼aAdvisor: Booth, Victoria;Schnell, Santiago.
■5021 ▼aThesis (Ph.D.)--University of Michigan, 2025.
■520 ▼aReduced models for enzymatic mechanisms, widely used in real-world biochemical applications, are often governed by parametric restrictions that determine their accuracy and applicability. The work in this thesis focuses on addressing challenges in analyzing solutions of reduced enzyme kinetic models, particularly in settings where assumptions of traditional analytic methods may fail. A second focus of this thesis is the development of a gradient-based optimization framework for fitting noisy differential equation models for the neuronal regulation of sleep-wake behavior to biological data.In Chapter 2, we derive slow-scale linear noise approximation (ssLNA) to stochastic enzyme kinetic models directly using geometric singular perturbation theory (GSPT). We resolve discrepancies among prior stochastic reductions by identifying the role of the Tikhonov standard form in producing consistent reductions. Extending the ssLNA to systems with a transcritical bifurcation, we disprove the widely-used Segel-Slemrod condition for stochastic model reduction in the Michaelis-Menten (MM) mechanism. Instead, we demonstrate that the more restrictive Reich-Sel'kov condition is needed, aligning stochastic and deterministic validity thresholds for quasi-steady-state approximations (QSSAs).In Chapter 3, we focus on the MM model to investigate the parametric conditions under which the standard QSSA is not only valid but also predominant among competing reductions. Using the theory of fences and anti-funnels, we locate the slow manifold in the MM phase plane and derive sharp error bounds. While the Reich-Sel'kov condition governs accuracy, we introduce a new and more restrictive qualifier that predicts the predominance of the standard QSSA. This distinction is critical when multiple reduced models are valid, as it guides optimal model selection for experimental design.In Chapter 4, we perform model reduction analysis for the Intermolecular Autocatalytic Zymogen Activation (IAZA) model, when it features a dynamic transcritical bifurcation. GSPT fails in this setting due to loss of normal hyperbolicity. Using the blow-up method, we transform the system near the bifurcation to recover hyperbolicity and derive an explicit solution to the desingularized system at the bifurcation point. This allows us to estimate the peak complex concentration analytically and explain the surprising accuracy of the nullcline-based reduction. We also identify the existence of a canard and analyze the implications of dynamic imperfections for model reduction.Chapter 5 tackles perturbation systems that lack an explicit Tikhonov standard form structure but mimic typical standard form behavior portraying fast convergence to QSS approximations. We propose a novel use of the blow-up method to connect the fast and slow dynamics in systems where a clear distinction is not apparent. Applied to the MM and IAZA systems, this approach reveals that the systems admit latent standard fast-slow structure and the trajectories starting both on and away from the critical manifold get attracted to the QSS curves, validating the resemblance to standard form dynamics.Chapter 6 shifts focus to computational neuroscience and presents a framework for fitting noisy differential equation models to rodent sleep-wake data using gradient-based optimization. We adapt a mechanistic model of interacting neuronal populations with noisy firing activity. Leveraging automatic differentiation, maximum mean discrepancy, and sensitivity analysis, we demonstrate efficient parameter inference from discrete sleep score data. The methodology offers a framework to relate experimental hypotheses for the neuronal circuit structures responsible for sleep-wake regulation to the random polyphasic sleep-wake behavior of rodents, with applications in and beyond sleep modeling.
■590 ▼aSchool code: 0127.
■650 4▼aBiology
■650 4▼aMathematics
■650 4▼aApplied mathematics
■650 4▼aBiochemistry
■653 ▼aMathematical biology
■653 ▼aGeometric singular perturbation theory
■653 ▼aQuasi-steady-state approximations
■653 ▼aSleep-wake regulation
■653 ▼aGradient-based optimization
■653 ▼aOrdinary differential equation
■690 ▼a0364
■690 ▼a0405
■690 ▼a0306
■690 ▼a0487
■71020▼aUniversity of Michigan▼bApplied and Interdisciplinary Mathematics.
■7730 ▼tDissertations Abstracts International▼g87-02B.
■790 ▼a0127
■791 ▼aPh.D.
■792 ▼a2025
■793 ▼aEnglish
■85640▼uhttp://www.riss.kr/pdu/ddodLink.do?id=T17359846▼nKERIS▼z이 자료의 원문은 한국교육학술정보원에서 제공합니다.


