GenePhLo: Genetic Phase-based Logic

就像电子计算机可以接受输入并使用算法来获得输出一样，活细胞也可以对一些输入分子做出反应并产生另一个作为回应。细胞没有电路，而是有一个相互联系的生物过程网络，在这个网络的基础上，我们有中心法则- DNA转录成RNA，RNA翻译成蛋白质，蛋白质可以调节DNA转录（Brophy和Voigt，2014）。利用已知功能的DNA，我们可以组装遗传逻辑电路。电子和生物计算最常见的方法是基于水平的：当信号（电压或生物分子浓度）低于一个阈值时，信号被计为低，或二进制0，当它高于另一个阈值时，它被计为高，或二进制1。虽然这种计算方法很普遍，甚至可以在自然系统中找到，如Ca 2+信号（Berridge，Lipp和Bootman，2000），但它对噪声不是很鲁棒。这在固有噪声的生物系统中可能是一个问题。例如，在相同的生长条件下，相同的蛋白质可以在细菌菌落中以不同的方式表达（比尔，2017），这可能导致计算结果冲突。此外，在基于电平的逻辑中，阈值之间的信号不是确定的，既不是0也不是1。编码二进制逻辑的另一种方式是使用振荡信号的相位。当信号与参考同相振荡时，这被计为二进制1，而当它异相时，则为0。分谐波注入锁定（SHIL）可用于使振荡信号振荡，以便相位可以响应于输入而移动（Roychowdhury，2015），记录单个信息位。为了达到这样的双稳态，需要在系统中加入两倍高频率的锁定信号。虽然基于相位的计算机对噪声更具鲁棒性，但这种方法尚未应用于生物计算。我想换掉这个。在我的项目中，我的目标是构建和表征基本的遗传相位逻辑电路。目标：表征由于信号分子的转录，翻译和扩散引起的延迟引起的相移。构建和表征基于相位的非门。比较基于相位和基于电平的非门的鲁棒性。构建和测试基于相位的互补MAJORITY，NAND和NOR门。使用Danino et al.（2010）振荡器测试SHIL。在我的项目中，我将应用计算和实验室方法。我将使用建模来探索一系列参数，并确定合适的实验设置。在实验室中，我将使用特征化的工程E。大肠杆菌细胞，并在微流控装置中培养它们，使用尼康Ti-E显微镜记录它们的行为。在需要的时候，我会编辑现有的基因电路或使用吉布森和金门组装方法设计新的。我将使用小信号分子和群体感应作为输入和“电线”，将具有不同遗传电路的细胞连接在一起。最后，我将把含有逻辑门的细胞连接到Danino等人（2010）的同步遗传振荡器菌落上，该菌落将充当状态寄存器。参考文献：比尔，J.（2017）“生物化学复杂性驱动遗传表达的对数正态变异”，工程生物学，1（1）。可在以下网址获得：https://doi.org/10.1049/enb.2017.0004。Berridge，M.J.，Lipp，P.和Bootman，M.D.（2000）“钙信号的多功能性和普遍性”，自然评论分子细胞生物学。网址：https：//doi.org/10.1038/35036035.Brophy，J.A.N.和Voigt，C.A.（2014）“遗传电路设计原理”，自然方法。网址：https：//doi.org/10.1038/nmeth.2926.Danino，T.等人（2010）“A synchronized quorum of genetic clocks”，Nature 2010 463：7279，463（7279），pp. 326-330.可从以下网址获取：https：//doi.org/10.1038/nature08753.Roychowdhury，J.（2015）“使用自维持非线性振荡器的布尔计算”，IEEE会议记录，103（11）。网址：https：

Just like electronic computers can take input and employ algorithms to get an output, so can living cells react to some input molecule and produce another in response. Instead of electric circuits, cells have a network of linked biological processes, at the base of which we have the central dogma - DNA is transcribed into RNA, RNA is translated into proteins, and proteins can regulate DNA transcription (Brophy and Voigt, 2014). Using DNA with known functions, we can assemble genetic logical circuits.The most common approach to both electronic and bio-computing is level-based: when a signal (electric voltage or a biomolecule concentration) is under one threshold, the signal is counted as low, or binary 0, and when it is above another, it is counted as high, or binary 1. While this computation method is widespread and can even be found in natural systems like Ca2+ signalling (Berridge, Lipp and Bootman, 2000), it is not very robust to noise. This can be a problem in intrinsically noisy biological systems. For example, the same protein can be expressed differently across the bacterial colony under the same growing conditions (Beal, 2017), which could result in conflicting computing outcome. What is more, in level-based logic the signal between the thresholds is not determined, neither 0 nor 1.Another way to encode binary logic is to use phases of oscillating signals. When a signal oscillates in phase with the reference, this is counted as binary 1, and when it is out of phase, 0. Sub-harmonic injection locking (SHIL) can be used to make oscillatory signal bistable, so that the phase could be shifted in response to an input (Roychowdhury, 2015), recording a single bit of information. To reach such bistability, a locking signal of twice higher frequency needs to be added to the system.While phase-based computers are more robust to noise, this approach has not yet been applied in biocomputing. I would like to change this. In my project, I aim to build and characterize basic genetic phase-based logic circuits.Objectives:Characterize phase shifts due to the delays caused by transcription, translation, and diffusion of signalling molecules.Build and characterize a phase-based NOT gate.Compare robustness of phase-based and level-based NOT gates.Build and test phase-based complimentary MAJORITY, NAND and NOR gates.Use Danino et al. (2010) oscillator to test SHIL.In my project, I will apply both computational and lab-based approaches. I will use modelling to explore a range of parameters and determine suitable experimental setups. In the lab, I will use characterized engineered E. coli cells and grow them in microfluidic devices, using a Nikon Ti-E microscope to record their behaviour. When needed, I will edit existing genetic circuits or engineer new using Gibson and Golden Gate assembly methods. I will use small signalling molecules and quorum sensing as inputs and "wires", linking cells with different genetic circuits together. Finally, I will connect cells containing logic gates to a synchronized genetic oscillator colony by Danino et al. (2010) which will act as a state register.References:Beal, J. (2017) 'Biochemical complexity drives log-normal variation in genetic expression', Engineering Biology, 1(1). Available at: https://doi.org/10.1049/enb.2017.0004. Berridge, M.J., Lipp, P. and Bootman, M.D. (2000) 'The versatility and universality of calcium signalling', Nature Reviews Molecular Cell Biology. Available at: https://doi.org/10.1038/35036035.Brophy, J.A.N. and Voigt, C.A. (2014) 'Principles of genetic circuit design', Nature Methods. Available at: https://doi.org/10.1038/nmeth.2926.Danino, T. et al. (2010) 'A synchronized quorum of genetic clocks', Nature 2010 463:7279, 463(7279), pp. 326-330. Available at: https://doi.org/10.1038/nature08753.Roychowdhury, J. (2015) 'Boolean Computation Using Self-Sustaining Nonlinear Oscillators', Proceedings of the IEEE, 103(11). Available at: https: