AitF: EXPL: Algorithmic Fluid Concentration Management for Programmable Microfluidics

技术的小型化不仅产生了可编程微处理器电路，而且还产生了用于微流体的可编程芯片实验室（PLoC）系统。 这些系统控制微小的液滴，允许精确的混合和测试，可以开发成环境监测器，疟疾诊断或癌细胞的血液测试。 与微处理器一样，在芯片上编程实验室需要了解芯片能做什么和不能做什么，以及一种合适的语言来表达命令。 这个项目1）发展了一个计算复杂性的理论，用于设计微流体装置，产生具有所需浓度的流体混合物，2）创建了一个特定领域的语言（包括应用程序接口或API和编译器），3）制造和测试设备，以及4）探索对包括流体浓缩步骤的复杂生化反应进行编程的能力的更广泛影响。项目成果将使PLoC系统更易于使用，PI将首先向学生（包括代表性不足的群体）展示，然后通过在科学会议上提供的教程，在更广泛的科学界推广PLoC技术的使用。该项目研究算法的复杂性和可近似性，以生成具有所需浓度的一个或多个离散流体液滴，例如，稀释至10%、15%等的样品，其目标是例如使稀释步骤的数量最小化、使浪费最小化、使反应物使用最小化等。这些算法将被推广以产生具有期望的时间依赖性浓度的液滴流。与此同时，将现场设计和制作一系列具有代表性的PLOC，沿着特定领域的语言和编译器，这将允许使用高级软件对PLOC进行编程。在这个项目的过程中开发的算法将被集成到一个API，可以从特定领域的语言调用。这将使用户能够指定包括流体浓度生成子程序（自动执行）的生化反应。将通过使用现场制造的PLoC执行这些生化反应来验证系统。在项目成功完成后，PLoC设计文件沿着连同特定领域的语言规范和编译器、文档和教程材料将公开发布，以鼓励其他科学家使用这个平台和方法来提高生产力，减少各自实验室的人为错误。

The miniaturization of technology has produced not only programmable microprocessor circuits, but also programmable lab-on-a-chip (PLoC) systems for microfluidics. These systems control tiny fluid droplets, allowing precise mixing and testing, that can be developed into environmental monitors, malaria diagnostics, or blood tests for cancer cells. As with microprocessors, programming a lab on a chip takes knowledge of what the chip can and cannot do, and a suitable language in which to express the commands. This project 1) develops a theory of computational complexity for designing microfluidic devices that produce fluid mixtures with desired concentrations, 2) creates a domain-specific language (including an Application Program Interface, or API and a compiler) for these devices, 3) fabricates and tests devices, and 4) explores the broader impacts of the ability to program complex biochemical reactions that include fluid concentration steps. The project results will make PLoC systems much easier to use, which the PIs will first demonstrate with students, including those from underrepresented groups, and then through tutorials presented at scientific meetings to promote the use of PLoC technologies among the broader scientific community. This project investigates the complexity and approximability of algorithms to generate one or more discrete fluid droplets having a desired concentration, e.g., a sample diluted to 10%, 15%, etc., with objectives such as minimizing the number of dilution steps, minimizing waste, minimizing reactant usage, etc. These algorithms will be generalized to produce of a stream of droplets having desired time-dependent concentrations. Meanwhile, a collection of representative PLoCs will be designed and fabricated on-site, along with a domain-specific language and compiler, which will allow the PLoCs to be programmed using high-level software. The algorithms developed in the course of this project will be integrated into an API that can be called from the domain-specific language. This will enable users to specify biochemical reactions that include fluid concentration generation subroutines (which are performed automatically). The system will be validated by executing these biochemical reactions using the PLoCs that have been fabricated on-site. Upon successful completion of the project, PLoC design files along with the domain-specific language specification and compiler, documentation, and tutorial material, will be released publicly to encourage other scientists to use this platform and approach to increase productivity and reduce human error in their respective laboratories.