EAGER: (ST1) Collaborative Research: Exploring the emergence of peptide-based compartments through iterative machine learning, molecular modeling, and cell-free protein synthesis

摘要：细胞是所有生物的基本组成部分。细胞内的物质被由脂肪酸分子组成的细胞膜分离并与环境隔离。这些分子在相对温和的自然条件下也能很好地发挥作用，比如在室温和常压下的水中。在恶劣的环境下，比如工业过程中普遍存在的极端温度和压力，这些分子是不稳定的，因此很难在这样的环境中部署细胞。此外，很难设计具有新功能的天然膜分子，如感知环境或结合特定表面或目标分子的能力。这项工作的主要目标是发现和合成一类基于蛋白质的新分子，作为合成细胞膜的替代“底盘”材料，这种材料提高了机械和化学稳定性，并可以通过工程设计赋予细胞膜新的功能。教授。Ferguson和Liu将快速实验合成和测试与计算机模拟和人工智能工具相结合，以寻找能够在恶劣环境中生存并能够将合成细胞组装成合成组织的新型合成膜。通过将实验与计算相结合，形成计算指导实验、实验指导计算建模的良性循环，与传统的试改进实验相比，大大节省了人工、时间和资源。在这项工作的过程中，教授。Ferguson和Liu为博士后、研究生、本科生和高中实习生提供研究机会，将研究成果纳入他们的教学课程，并通过科学与工程女孩夏令营、底特律地区大学预科工程项目和芝加哥大学课余事务暑期实习项目参与推广活动。技术摘要：这项工作的目的是发现新的肽类生物材料作为合成细胞的替代“底盘”材料。虽然生物学已经确定使用脂质双层膜作为分隔细胞质和膜结合细胞器的材料，但多肽提供了一种替代生物材料，可以建立具有改进的机械和化学稳定性以及额外工程生物功能的肽微室。与脂质和聚合体膜材料相比，肽基基底材料在生物相容性、化学和机械稳定性以及额外功能化能力方面具有独特的优势，这使得它们在生物医学、药物输送、生物传感和非自然环境中的部署方面非常理想。能够自发自组装到溶质填充的具有所需材料特性的微室中的肽序列的发现受到蛋白质序列搜索空间的巨大尺寸的挫折，这使得详尽的探索变得棘手，爱迪生试验和改进效率低下。在这项工作中，我们建立了一个集成的数据驱动建模和高通量无细胞合成平台，以快速遍历序列空间。在理论和建模之间紧密集成的反馈回路中，我们采用“主动学习”范式从实验数据中提取信息，指导对巨大肽序列空间的合理遍历，并优化配置实验资源。这项工作的完成将导致发现备受追捧的基于肽的合成细胞“底盘材料”，这种材料对恶劣环境更加坚固，并且具有互补的结合功能，使合成细胞能够自组织成合成组织。在这项工作的过程中，教授。Ferguson和Liu为高中生和本科生提供了研究机会，让他们接触科学研究，提高STEM项目的代表性，培训研究生和博士后学员进行综合计算和实验研究，将科学成果纳入课程材料，并通过科学与工程女孩夏令营、底特律地区大学预科工程项目参与外展活动。以及芝加哥大学课余事务暑期实习项目。该奖项反映了美国国家科学基金会的法定使命，并通过使用基金会的知识价值和更广泛的影响审查标准进行评估，被认为值得支持。

Non-technical Abstract:The cell is the fundamental building block of all living things. The materials within the cell are separated and protected from the environment by the cell membrane that is composed of molecules derived from fatty acids. These molecules function well under relatively benign natural conditions, such as in water at room temperature and pressure. Under harsh environments, such as extreme temperatures and pressures prevalent in industrial processes, these molecules are unstable therefore making it difficult to deploy cells within such environments. Furthermore, it is difficult to engineer natural membrane molecules with new functions such as the ability to sense their environment or bind particular surfaces or target molecules. It is the primary goal of this work to discover and synthesize a new class of molecules based on proteins as an alternative 'chassis' material for synthetic cell membranes that have improved mechanical and chemical stability and can be engineered to endow the membrane with new functions. Profs. Ferguson and Liu combine fast experimental synthesis and testing with computer simulations and artificial intelligence tools to search for new synthetic membranes with the ability to survive in harsh environments and the capacity to assemble synthetic cells together into synthetic tissues. By combining experiment and computation within a virtuous cycle, wherein computation guides experiment and experiment informs computational modeling, massive savings in labor, time, and resources are realized compared to traditional trial-and-improvement experimentation. In the course of this work, Profs. Ferguson and Liu provide research opportunities for post-doctoral, graduate, undergraduate, and high-school trainees, incorporate the outcomes of the research into classes that they teach, and engage in outreach activities through a Girls in Science and Engineering summer camp, Detroit Area Pre-College Engineering Program, and University of Chicago After School Matters summer internship program.Technical Abstract:The aim of this work is to discover novel peptidic biomaterials as an alternative "chassis" material for synthetic cells. While biology has settled on using lipid bilayer membrane as the material for compartmentalizing cytoplasm and for membrane-bound organelles, polypeptides offer an alternative biomaterial that can establish peptidic microcompartments with improved mechanical and chemical stability and the capacity for additional engineered biological function. Peptidic chassis materials offer unique advantages compared to lipid and polymersome membrane materials in terms of biocompatibility, chemical and mechanical stability, and capacity for additional functionalization that make them extremely desirable for applications in biomedicine, drug delivery, biosensing, and deployment in non-natural environments. The discovery of peptide sequences capable of spontaneous self-assembly into solute-filled microcompartments with desired materials properties is frustrated by the vast size of the protein sequence search space that makes exhaustive exploration intractable and Edisonian trial-and-improvement inefficient. In this work, we establish an integrated data-driven modeling and high-throughput cell-free synthesis platform to rapidly traverse sequence space. In a tightly integrated feedback loop between theory and modeling, we employ an 'active learning' paradigm to extract information from experimental data, guide rational traversal of the vast peptide sequence space, and optimally deploy experimental resources. Completion of this work will lead to the discovery of highly sought-after peptide-based synthetic cell 'chassis materials' that are more robust to harsh environments and which have complementary binding functionality to enable self-organization of the synthetic cells into synthetic tissues. In the course of this work, Profs. Ferguson and Liu offer research opportunities for high-school and undergraduate students to provide exposure to scientific research and improve representation in the STEM pipeline, train graduate and post-doctoral trainees in integrated computational and experimental research, incorporate the scientific outcomes in course materials, and engage in outreach activities through a Girls in Science and Engineering summer camp, Detroit Area Pre-College Engineering Program, and University of Chicago After School Matters summer internship program.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.