CAREER: Neuronal Data Assimilation Tools and Models for Understanding Circadian Rhythms

人类拥有一种称为生物钟的内部计时机制，它使生理过程与一天中的适当时间保持一致，例如通过在晚上刺激促进睡眠的激素褪黑激素的释放和在早上刺激促进唤醒的激素皮质醇的释放。在跨时区快速旅行后，生物钟与24小时环境周期的失调会发生，并导致时差症状，包括睡眠障碍，消化问题和认知能力下降。生物钟在动物物种中是高度保守的，人类和果蝇的生物钟系统所涉及的基因和神经回路有许多相似之处。该项目的目标是通过开发果蝇生物钟的详细模型，然后系统地分析生物钟如何被外部光暗周期时间的突然变化所破坏，从而从数学上理解时差反应的生物学基础。计算模型的预测将通过模拟跨子午线旅行的实验在苍蝇中进行测试。该项目还将创建新的工具，通过汇集应用数学，统计学和神经科学的技术，直接从观测数据中构建数学模型。将在这些领域的交叉点开发一个独特的研究生课程。研究生，本科生和社区学院的学生将参与这项研究，并获得跨学科的培训。本科生和社区大学生将组成综合暑期研究项目团队。社区大学的学生将被招募参加新泽西理工学院的本科数学建模课程，并将获得四年制STEM学位课程的指导机会。该项目侧重于神经元建模背景下动力系统和统计数据分析之间的接口。该项目的研究目标是开发新的数据同化方法，用于直接从时间过程数据推断神经元动力学模型，从而深入了解生物机制。具体而言，该项目将创建新的数据同化工具，以识别和参数化来自果蝇昼夜节律（~24小时）时钟网络记录的电压轨迹的神经生理模型。这项研究工作将解决数学/计算神经科学领域的两个主要空白：缺乏从单独测量膜电压推断未观察到的离子电流参数的方法，以及缺乏将分子，细胞和行为尺度联系起来的模型。这将通过开发一种方法来设计用于数据同化算法的刺激协议来实现，该算法最佳地揭示了不可观察的神经元变量的动态，并改善了电生理模型和参数的推断。该方法将用于建立果蝇时钟网络模型，该模型将基因表达中的昼夜节律与神经元活动和行为输出的变化联系起来。这些模型将用于分析内部昼夜节律振荡器如何在外部明暗周期中的相移之后重新夹带。更一般地说，这项工作的目的是阐明振荡系统中的同步和夹带的基本方面。

Human beings possess an internal timekeeping mechanism known as the circadian clock that aligns physiological processes with the appropriate time of day, for example by stimulating the release of the sleep-promoting hormone melatonin in the evening and the wake-promoting hormone cortisol in the morning. Misalignment of the circadian clock with respect to 24-hour environmental cycles occurs after rapid travel across time zones and leads to symptoms of jet lag including sleep disturbances, digestive problems, and decreased cognitive performance. The circadian clock is highly conserved across animal species, and there are many similarities in the genes and neural circuits involved in the circadian systems of humans and the fruit fly Drosophila. The goal of this project is to obtain a mathematical understanding of the biology underlying jet lag by developing a detailed model of the circadian clock in Drosophila and then systematically analyzing how the clock is disrupted by sudden changes in the timing of the external light-dark cycle. The predictions of the computational model will then be tested in flies through experiments that simulate transmeridian travel. This project will also create new tools for building mathematical models directly from observed data by bringing together techniques from applied mathematics, statistics, and neuroscience. A unique graduate-level course at the intersection of these fields will be developed. Graduate, undergraduate, and community college students will be involved in this research and obtain interdisciplinary training. The undergraduate and community college students will form integrated summer research project teams. Community college students will be recruited to enroll in an undergraduate-level mathematical modeling course at the New Jersey Institute of Technology and will receive mentorship on pursuing four-year STEM degree program opportunities.This project focuses on the interface between dynamical systems and statistical data analysis in the context of neuronal modeling. The research goal of this project is to develop novel data assimilation methodologies for inferring models of neuronal dynamics directly from time-course data that enable insights into biological mechanisms. Specifically, this project will create new data assimilation tools to identify and parameterize neurophysiological models from voltage traces recorded from the Drosophila circadian (~24-hour) clock network. This research effort will address two major gaps in the mathematical/computational neuroscience field: a scarcity of methods for inferring parameters of unobserved ionic currents from measurements of membrane voltage alone, and a lack of models that link molecular, cellular, and behavioral scales. This will be accomplished by developing a method to design stimulus protocols for use in data assimilation algorithms that optimally unmask the dynamics of unobservable neuronal variables and improve inference of electrophysiological models and parameters. The method will be used to build a model of the Drosophila clock network that links circadian rhythms in gene expression to changes in neuronal activity and behavioral outputs. These models will be used to analyze how the internal circadian oscillator re-entrains following phase shifts in the external light-dark cycle. More generally, this work aims to elucidate fundamental aspects of synchronization and entrainment in oscillatory systems.