Synthetic cells that can learn without evolution

适应是生物学的基本特征。原则上，生命系统可以通过两种机制进行适应：进化和学习。学习是一种潜在的快速而有力的适应方式。即使是最简单的细胞也能进化，但它们能在没有进化的情况下展示学习能力吗？如果是这样的话，它们能参与什么样的学习模式，学习细胞或类似细胞的系统能有多简单？在着手解决这些关于生命法则的基本问题时，这个项目将有助于定义学习系统的基本生物学性质。该项目将在自下而上构建“智能”合成细胞系统方面取得重要进展，未来可能在广泛的学术、工业、临床和环境环境中应用。将招收多学科的研究生，训练他们进行跨学科研究，并为生物工程相关课程开发一套“科学与社会”模块。此外，该项目将通过开发与项目目标相关的实践活动吸引更广泛的公众受众。该项目旨在创造能够联想学习的合成细胞系统。具体来说，该项目将开发一种合成细胞，该细胞通过将光脉冲信号与嗅觉受体检测到的分子添加相关联，从而学会对光脉冲信号做出反应。成功将提供一个原理证明，遗传编码信息处理系统可以执行学习任务，并将产生一个可重复使用的学习电路基序库。联想学习电路的建模和设计将为相应基因调控电路结构的发展提供信息。利用嗅觉受体蛋白库来感知多输入化学信号，并研究膜封装对系统行为的影响。最后，综合人类实践部分将从历史、概念和伦理的角度探讨学习合成细胞和人工神经网络/机器学习之间的关系。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。

Adaptation is a fundamental and defining feature of biology. In principle, living systems can adapt via two mechanisms: evolution and learning. Learning is a potentially rapid and powerful mode of adaptation. Even the simplest cells can evolve, but can they demonstrate learning in the absence of evolution? If so, what modes of learning can they engage in, and how simple can learning cells or cell-like systems be? In setting out to address these fundamental questions about the Rules of Life, this project will help to define the essential biological nature of learning systems. This project will make important progress towards the bottom-up construction of 'smart' synthetic cell systems, with potential future applications across a wide range of academic, industry, clinical and environmental settings. A multi-disciplinary cohort of graduate students will be recruited and trained in interdisciplinary research, and a set of 'science & society' modules for bioengineering-related courses will be developed. Furthermore, the project will engage a broader public audience by developing hands-on activities related to the goals of the project.This project aims to create synthetic cell systems capable of associative learning. Specifically, the project will develop a synthetic cell that learns to respond to a light pulse signal by associating it with the addition of molecules detected by olfactory receptors. Success will provide a proof-of-principle that genetically encoded information-processing systems can carry out learning tasks, and will generate a reusable library of learning circuit motifs. Modeling and design of associative learning circuits will inform the development of corresponding genetic regulatory circuit architectures. Multi-input chemical signals will be sensed using a library of olfactory receptor proteins, and the effects of membrane encapsulation on system behavior will be studied. Finally, an integrated Human Practices component will explore the relationship between learning synthetic cells and artificial neural networks/machine learning, from historical, conceptual and ethical perspectives.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.