SemiSynBio: Collaborative Research: Very Large-Scale Genetic Circuit Design Automation

生物学的计算能力是令人难以置信的，在自然界中，材料和动物的身体结构的复杂模式都是显而易见的。细胞通过使用相互作用的生物分子网络来构建这些结构，这些生物分子编码在它们的DNA中，其功能就像微型计算机，当许多细胞通过交流来共同解决一个问题时，其能力就会增长。该项目的目标是通过设计显著扩大构建这些系统的能力，这样细胞就可以被编程来执行复杂的计算任务。这将通过创建允许用户编写代码的软件来实现，就像给计算机编程一样，然后将其编译成DNA序列。新的理论工具将用于确定细胞运行这些程序所需的功率，以及如何最好地在细胞编码电路和传统电子系统之间分配任务。这项研究将广泛影响生物技术，生物技术正越来越多地被用于商业生产各种产品，从消费品到高端先进材料。目前的产品没有利用细胞的计算潜力；换句话说，所有的基因都一直处于开启状态。这项研究将使细胞能够通过在细胞内部进行计算和在细胞之间进行交流，通过多个步骤来构建化学物质和材料。这项工作是跨学科的，需要生物学、化学、数学、生物工程、电子工程和计算机科学的背景。因此，该项目包括开发新的教育平台，以满足行业对在传统分离领域之间接受培训的学生的需求。这包括一个新的本科合成生物学设计课程，一个工业合作社，以及课程材料“如何种植几乎任何东西”，这将在国际层面上公开。为了构建自然世界的复杂性，细胞使用由相互作用的生物分子组成的调节网络来控制基因调节的时间和条件。在过去的20年里，研究人员已经能够通过巧妙地结合调控相互作用来构建合成遗传电路。问题是，这种电路中最大的只有大约10个调节器，比自然网络小得多，这极大地限制了可以执行的计算。拟议的研究将开发技术，这些技术共同使计算复杂性大规模扩展到~10^5个调节器。第一个目标是增加细胞内回路的大小。基于Cas9的逻辑门具有巨大的扩展潜力，但受到dCas9毒性和序列重复的限制。在数学建模的指导下，将设计一组门来解决这些问题。设计自动化的框架将被开发出来，使Verilog规范能够转换成逻辑图，然后在许多相互作用的单元之间进行划分。第二个目标是在多个通信细胞之间分配遗传电路设计。细胞间通信信号的数量和可靠性将通过定向进化从2个通道增加到8个通道来提高。这些将在活细胞和非活系统中实施，从而实现生物反应器内外的广泛应用。结合50门/单元，该平台提供了包含10^5+门的多细胞电路的可能性。一些应用程序需要作为非活动系统部署，例如当应用程序在实验室之外时，因此需要包含。第三个目标是将目标1和2中开发的部分转化为包封无细胞蛋白提取物的多个通信脂质囊泡。Cas9门和额外的通信通道将被表征，以扩大计算潜力。这些将被描述为门，并使用电子设计自动化工具实现大型系统的自动化设计。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。

The computing power of biology is incredible, evident in the natural world in the intricate patterns underlying materials and the body plan of animals. Cells build these structures by using networks of interacting bio-molecules, encoded in their DNA, that function as microscopic computers, the power of which grows as many cells communicate to work together on a problem. The goal of this project is to significantly scale-up the ability to build these systems by design such that cells can be programmed to perform complex computational tasks. This will be done by creating software that allows a user to write code, exactly as one would program a computer, which is then compiled to a DNA sequence. New theoretical tools will be applied to determine the power required by the cell to run these programs and how best to distribute tasks between circuits encoded in cells and conventional electronic systems. This research will broadly impact biotechnology, which is increasingly being used to commercially produce a wide range of products, from consumer goods to high-end advanced materials. Current products do not harness the computational potential of cells; in other words, all the genes are turned on all the time. This research will enable cells to be programmed to build chemicals and materials in multiple steps, both by performing the computations inside of the cells and also communicating across cells. This work is interdisciplinary and requires backgrounds in Biology, Chemistry, Mathematics, Biological Engineering, Electrical Engineering, and Computer Science. As such, the project includes the development of new educational platforms in anticipation of a need in industry for students trained at the interface between traditionally separated fields. This includes a new undergraduate-level Synthetic Biology Design course, an industrial co-op, and curriculum material "How to Grow Almost Anything," which will be made public at an international level. To build the complexity of the natural world, cells use regulatory networks made up of interacting bio-molecules to control the timing and conditions for gene regulation. For the last 20 years, researchers have been able to build synthetic genetic circuits by artfully combining regulatory interactions. The problem is that the largest of such circuits only consist of ~10 regulators, far smaller than natural networks, which drastically limits the computation that can be performed. The proposed research will develop technologies that collectively enable a massive scale-up in computational complexity to ~10^5 regulators. The first objective seeks to increase the size of circuits within cells. Logic gates based on Cas9 have enormous scale-up potential, but are limited by dCas9 toxicity and sequence repeats. A set of gates will be designed to fix these problems, guided by mathematical modeling. A framework for design automation will be developed that enables a Verilog specification to be converted into a logic diagram, that is then divided up amongst many interacting cells. The second objective seeks to distribute a genetic circuit design across multiple communicating cells. The number and reliability of cell-cell communication signals will be improved by directed evolution to increase the number of channels from 2 to 8. These will be implemented in living cells and non-living systems, thus enabling a broad range of applications inside and outside the bioreactor. Combined with 50 gates/cell, this platform offers the possibility of multicellular circuits containing 10^5+ gates. Some applications require deployment as a non-living system, for example when the application is outside of the lab, thus requiring containment. The third objective seeks to translate the parts developed in Objectives 1 and 2 to operate in multiple communicating lipid vesicles encapsulating cell-free protein extract. Cas9 gates and additional communication channels will be characterized to expand the computational potential. These will be characterized as gates and implemented using Electronic Design Automation tools to automate the design of large systems.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.