WHole Animal Modelling (WHAM): Toward the integrated understanding of sensory motor control in C. elegans

动物是了不起的生物。任何人造机器在导航复杂环境、对丰富的感官线索做出反应、在遇到全新场景时学习和调整行为等方面都无法与之匹敌。但即使是最简单的动物，即使是最简单的神经系统也能做到这一点。简单的动物可能不会下国际象棋或平衡银行对账单，但我们可以从它们那里学到很多关于它们强大的感觉整合和运动控制机制的知识，这可能对我们在控制工程、生物机器人甚至未来正在开发的神经假肢应用的脑机接口中有用。理解动物行为的一个很好的起点是一种微小的、自由生活的1毫米长的线虫，称为线虫。与我们的1000亿个神经细胞相比，甚至与一只苍蝇的10万个神经细胞相比，这种蠕虫的整个神经系统只有302个神经细胞。与我们的神经系统不同，蠕虫的电路是硬连接的，在该物种的个体中是相同的，这使得严格和可重复地研究成为可能。部分由于它的简单性，部分由于它在实验室中易于操作，这是唯一一种整个神经系统被精确绘制(到亚细胞分辨率)的动物。但是，尽管它相对简单，这种蠕虫拥有许多被归因于更复杂的动物的功能，包括取食、交配、复杂的感觉能力、记忆和学习。因此，毫不奇怪，这种蠕虫的建模激发了物理学家、计算机科学家和工程师的想象力。线虫的综合建模甚至被提议作为英国的“计算研究的重大挑战”之一。在这个奖学金中，我将开始将我们对线虫感觉运动行为的理解整合到一个单一的计算模型中。挑战是弥合有效的静态神经电路结构和它所支持的动态神经计算之间的差距。这个奖学金将使我能够实现一步的改变，不仅在我们对重要的模型生物体的理解上，而且在推进复杂系统的科学和工程方面，无论是在逆向工程现实世界网络的背景下，还是在设计它们的背景下。

Animals are remarkable creatures. No man-made machine even comes close in its ability to navigate complex environments, respond to rich sensory cues, learn and adapt its behaviour when encountering completely novel scenarios and much much more. But even the simplest animals, ruled by even the simplest nervous systems, can achieve this. Simple animals may not be able to play chess or balance bank statements, but there is much we can learn from them about their robust mechanisms for sensory-integration and motor control which may be of use to us in control engineering, bio-robotics and even in future brain-machine interfaces that are being developed for neuro-prosthetic applications.An excellent starting point for understanding animal behaviour is a tiny, free living 1mm long roundworm, called C. elegans. By comparison to our 100 billion nerve cells, or even a fly's 100 thousand nerve cells, this worm's entire nervous system consists of a mere 302 nerve cells. Unlike our nervous system, the worm's circuitry is hard wired and identical across individuals of the species, making it possible to study rigorously and reproducibly. Due in part to its simplicity, and in part to its ease of manipulation in the lab, this is the only animal for which this entire nervous system has been mapped in exquisite detail (to sub-cellular resolution). But despite its relative simplicity, this worm possesses many of the functions that are attributed to more complex animals, including feeding, mating, complex sensory abilities, memory and learning. It is not surprising, therefore, that the modelling of this worm has captured the imagination of physicists, computer scientists and engineers alike. The integrated modelling of C. elegans has even been proposed as one of the UK's ``Grand Challenges for Computing Research.''In this Fellowship, I will begin to integrate our understanding of C. elegans sensory motor behaviour in a single computational model. The challenge is to bridge the gap between the effectively static neural circuit architecture and the dynamic neural computation it sustains. This fellowship will enable me to deliver a step change, not only in our understanding of an important model organism, but also in advancing the science and engineering of complex systems, whether in the context of reverse engineering real world networks, or in the context of designing them.