The Evolution of Dynamic Response Strategies: Optimal Control and Evolutionary Dynamics

智力优势：该项目将发展基于最优控制理论和群体遗传学的理论方法，以了解从单细胞到大型动物的生物如何感知环境的变化并动态地做出反应。为了生存，生物必须不断地对环境的变化做出反应，包括食物供应的变化以及捕食者的数量和类型的变化。例如，动物与同类的其他成员争夺食物；因此，其他个体的觅食策略会导致食物供应的随机波动。随着时间的推移，环境的动态变化包含了个体生物必须感知、处理、记忆和采取行动的信息。然而，由于不完整或不准确的感知和处理，这些内部过程受到噪声的影响。这在细胞水平和整个生物体水平上都是正确的，例如，当人类通过观察周围不断变化的环境来规划未来时。因为生物已经进化出了感觉处理策略来应对物种所经历的历史波动，环境波动方式的系统变化可能导致生物误解变化并表现出不适应。该项目考虑了遗传编码的动态响应策略，使生物体能够在给定的时间序列感知输入的情况下产生响应。这些动态响应策略可以考虑到感官输入的有限可靠性以及环境波动的频率和大小的变化。该项目重点研究细胞动态反应系统，包括简单基因调控、网络中相互作用基因的调控、细胞感觉机制和竞争性觅食。动态响应的机制细节将使用工程方法建模，进化结果将使用群体遗传技术确定。该项目将结合来自工程科学的强大计算技术和来自生物科学的种群进化限制的详细知识。工程解决方案将提供进化可以实现的动态响应策略改进范围的上限。这些限制将用于评估非最佳机制是否足以使生物体提取环境中可用的大部分资源。该项目具有潜在的变革性，因为所开发的方法有望广泛适用于生物尺度，并适用于发育、生理和行为反应。更广泛的影响：这个项目的一个更广泛的影响包括训练高中生的科学方法和工程原理。每年将招收四名就读于Dos Peublos高中工程学院的高中生，在赫斯潘哈的实验室参加暑期实习项目。这些学生已经参加了密集的工程课程，在那里他们学习微积分和计算机编程。学生实习生将学习使用iRobot Create平台对机器人进行编程，以执行合作和竞争任务。学生将学习运用科学的方法来改进算法的性能。学生们将把他们在暑期学到的技能应用到他们的学术机器人团队中。该项目的第二个方面是对加州大学圣迭戈分校的生物学本科生进行理论生物学培训。PI Proulx已经开发了一门使用Mathematica编程环境建模行为动力学的本科课程。资金将支持世界一流教师的访问，以丰富本课程，并根据他们自己目前的研究在生物系统数学建模方面举办研讨会，并为参与的学生提供访问专家对他们建模研究的即时反馈。这个项目最后一个更广泛的影响是为本科生提供暑期实习计划。该项目为加州大学圣巴巴拉分校的本科生和来自当地社区学院（包括圣巴巴拉城市学院）的少数民族学生提供了机会，让他们在夏季期间在共同负责人Hespanha及其研究生的指导下共同完成研究项目。参与的研究生将获得宝贵的指导经验。

Intellectual Merit: This project will develop theoretical approaches based on optimal control theory and population genetics to understand how living organisms, ranging in size from single cells to large animals, sense changes in their environments and respond to them dynamically. To survive, living organisms must constantly respond to changes in their surroundings, including changes in food supplies and the numbers and types of predators. For example, animals compete with other members of their species for food; thus the foraging strategies of other individuals cause random fluctuations in food availability. The dynamic changes in conditions over time contain information that individual organisms must sense, process, remember, and act on to their advantage. However, these internal processes are subject to noise due to incomplete or inaccurate sensing and processing. This is just as true at the cellular level as it is at the level of whole organisms, for example when humans plan for the future by observing the changing conditions around them. Because organisms have evolved sensory processing strategies in response to the historical fluctuations that the species has experienced, systematic changes in the way that the environment fluctuates can cause the organism to misinterpret change and behave maladaptively. This project considers the genetically encoded dynamic response strategies that enable living organisms to produce a response given a time series of sensed inputs. These dynamic response strategies can take into account the limited reliability of sensory input and changes in the frequency and size of environmental fluctuations. The project focuses on cellular dynamic response systems including simple gene regulation, regulation of interacting genes in a network, cellular sensory mechanisms, and competitive foraging. The mechanistic details of the dynamic response will be modeled using engineering methods and evolutionary outcomes will be determined using population genetic techniques. This project will combine powerful computational techniques from the engineering sciences with detailed knowledge of the constraints on evolving populations from the biological sciences. The engineering solutions will provide upper bounds on the range of improvement in dynamic response strategies that evolution can achieve. These limits will be used to evaluate whether non-optimal mechanisms are adequate for organisms to extract most of the resources available in the environment. This project is potentially transformative because the methods to be developed are expected to be broadly applicable across biological scales and to developmental, physiological, as well as behavioral responses.Broader Impact: One broader impact of this project involves training high school students in the scientific method and engineering principles. Four high school students enrolled in the Dos Peublos High School Engineering Academy will be recruited each year to participate in a summer internship program in co-PI Hespanha's lab. These students are already involved in an intensive engineering program where they learn calculus and computer programming. The student interns will learn to program robots to carry out cooperative and competitive tasks using use the iRobot Create platform. Students will learn to apply the scientific method to refine the performance of their algorithms. Students will apply the skills they learn in the summer as members of their scholastic robotics team.A second aspect of this project involves training undergraduate biology students from UCSB in theoretical biology. PI Proulx has already developed an undergraduate course on modeling behavioral dynamics using the Mathematica programming environment. Funding will support visits by world-class faculty to enrich this course with workshops in mathematical modeling of biological systems based on their own current research and provide participating students with immediate feedback on their modeling studies from visiting experts.The final broader impact of this project involves a Summer Internship Program for undergraduate students. This program provides opportunities for UCSB undergraduates and minority students from local community colleges, including Santa Barbara City College, to work together during the summer on research projects under the supervision of co-PI Hespanha and his graduate students. Participating graduate students will gain valuable mentoring experience.