CAREER: Stochastic Analyses to Optimize Designs for Single-Cell Optical Microscopy Experiments

计算分析可以为解开基本生物学之谜、发现新机制以及推动国家在医学、农业和环境科学方面的进步提供强有力的钥匙。生物过程的异常复杂的性质要求创建新的计算工具，以应对现代生物学研究中经常出现的特定挑战和机遇。该项目发现，改进和验证了几套集成的计算和实验方法，以探索和更全面地了解分子生物学中最基本的过程之一-单个mRNA分子的翻译形成功能蛋白质。该项目创建和分发新的计算软件包，以设计，解释和预测尖端的单细胞和单分子实验，用于未来的生物和生物医学研究。该项目还包括一个新的为期三周的年度暑期学校计划，以培训物理，工程和计算机科学的本科生，使用现代统计方法来测量，分析，预测和控制复杂的生物过程。此外，该项目还涉及为高年级本科生和研究生开发和传播在线定量生物学课程，以补充最近社区编写的教科书《定量生物学》：理论、计算方法和模型（麻省理工学院出版社，2018年）。该项目创建，测试和验证新技术，将超分辨率荧光显微镜实验与高性能计算相结合，解释和预测可变基因组和环境背景下单个mRNA翻译的复杂动态。项目目标包括三个综合研究目标：（1）开发严格的计算工具来量化和利用内在和外在波动（通常称为“过程噪声”），其影响单个mRNA翻译的动力学，同时最小化实验测量噪声的影响;（二）展示新的方法来严格估计模型的不确定性，并选择最佳的实验来解决竞争中的细微差异。以最有效和最具成本效益的方式进行假设;和（3）将联合收割机内在的时间mRNA翻译波动与统计分析相结合，以使得能够在真实的时间内和在单个活细胞中同时测量多个翻译mRNA种类。该项目整合了所有三个目标的工具，以生产和分发一个开源的，用户友好的Python软件包，称为RNA Sequence to NAScent Protein Simulator（rSNAPSim），它使实验生物学家能够分析，预测，重新设计一个mRNA翻译实验。该奖项反映了NSF的法定使命，并被认为值得通过使用基金会的智力价值和更广泛的影响进行评估来支持审查标准。

Computational analyses can provide powerful keys to unravel basic biological mysteries, to discover new mechanisms, and to drive national advances in medical, agricultural, and environmental sciences. The exceptionally complicated nature of biological processes requires that new computational tools be created to address specific challenges and opportunities that frequently arise in modern biological investigations. This project discovers, improves, and validates several sets of integrated computational and experimental methods to explore and more fully understand one of the most fundamental processes in molecular biology – the translation of individual mRNA molecules to form functional proteins. The project creates and distributes new computational software packages to design, interpret, and predict cutting-edge single-cell and single-molecule experiments for use in future biological and biomedical investigations. This project also incorporates a new three-week annual summer school program to train undergraduate students from the physical, engineering, and computer sciences, to use modern statistical methods to measure, analyze, predict, and control complex biological processes. In addition, the project involves developing and disseminating an online quantitative biology course curriculum for senior undergraduate and graduate students to complement the recent community-written textbook Quantitative Biology: Theory, Computational Methods, and Models (MIT Press, 2018).This project creates, tests, and validates new technologies to integrate super-resolution fluorescence microscopy experiments with high-performance computing to measure, interpret and predict the complex dynamics of single-mRNA translation within variable genomic and environmental contexts. Project goals comprise three integrated research objectives: (1) Develop rigorous computational tools to quantify and take advantage of intrinsic and extrinsic fluctuations (often called ‘process noise’) that affect the dynamics of single-mRNA translation, while minimizing impacts of experimental measurement noise; (2) Demonstrate new methodologies to rigorously estimate model uncertainties and choose the best possible experiments to resolve subtle differences in competing hypotheses in the most efficient and cost-effective manner; and (3) Combine intrinsic, temporal mRNA translation fluctuations with statistical analyses to enable simultaneous measurement of multiple translating mRNA species in real time and in single, living cells. The project incorporates tools from all three objectives to produce and distribute an open-source, user-friendly Python software package called RNA Sequence to NAscent Protein Simulator (rSNAPSim), which enables experimental biologists to analyze, predict, and redesign single-mRNA translation experiments.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.