Coding in healthy and diseased neurons

我们的神经系统从环境中收集信息，从而产生感知和适当的行为。关于如何获取和处理这些信息，还有一些令人兴奋的未解决的基本问题。这一建议使用数学来解决三个这样的基本问题。 1)温度传感器是如何工作的？研究“专家”是如何做到这一点的，可能会揭示所有感官的基本组织原则。我们将研究蛇如何能够探测到环境中极小的温度变化（例如由于猎物）。它们通过头上专门的红外摄像机测量热量。它们可以对千分之一度范围内的变化做出反应。80年代中期的工作表明，这种感觉在背景噪声允许的极限附近工作，这妨碍了信号检测。然而，从那时起，在热敏神经元中发现了被称为TRP受体的极端温度敏感的离子通道。在这些神经元中还发现了新的电同步现象。拟议的研究将建立，使用神经建模，如何重要的同步和TRP受体的热敏感性。这在从相机的红外光学到共同处理红外和视觉图像的更高大脑水平的链条中添加了一个基本环节。 2)神经损伤和信息流退化之间的关系是什么？神经是由单个神经元发出的被称为“轴突”的“电线”的集合。信息以电脉冲的形式沿着神经快速传播，依赖于髓磷脂。这种髓鞘是由包裹在轴突周围的专门细胞组成的，除了被称为朗维尔节点的微小1微米间隙。髓磷脂改变轴突“电缆”的电特性，使脉冲在节点之间快速跳跃。它还将一个轴突与另一个轴突隔离。髓鞘在多发性硬化症（MS）等疾病中变薄。最近，MS的损伤可以通过最先进的“汽车”显微镜以高分辨率进行量化。我在U的同事。拉瓦尔给我提供了一只老鼠在疾病不同阶段的图像。我们的工作建议从图像中了解和预测这些神经的功能缺陷。这涉及在健康和脱髓鞘的单个轴突和轴突束中建模传播。 3)刺激是如何改变大脑节律的？关于节奏是如何产生的以及它们的含义，争论激烈。某些脑节律被认为对我们注意刺激的能力很重要。我们的工作提出了一种简单的新机制，可以在弱电鱼脑中进行测试。巧合的是，我们预测节律出现在接收随机信号加上该信号延迟版本的细胞中（这种“前馈”电路比比皆是）。这将使用我们最近开发的网络中细胞活动的理论进行详细分析。这一理论也将被扩展到解释大脑节律如何与刺激保持一致，这是某些疾病中缺失的现象。 这项工作是理论和计算。我们制定了预测数学模型，这些模型足够简单，可以用最少的生物成分来解释神经现象。这项工作将仔细利用现有的生物学数据，并从与实验学家的合作中受益。建模还可以通过理论分析或计算机实验对功能进行预测，这些功能可以在新的实验中进行测试。这项研究将加深我们对脑回路功能的认识。它还可能导致基于新发现的原理的新技术（例如神经修复术），包括修复故障电路或感知热量的技术。

Our nervous system gathers information from the environment, resulting in perceptions and appropriate behaviours. There are exciting unresolved fundamental questions about how this information is acquired and processed. This proposal uses mathematics to address three such fundamental questions. 1) How do temperature sensors work? Studying how the "experts" do it may reveal fundamental organization principles for all senses. We will investigate how snakes can detect extremely small temperature changes in their environment (due e.g. to prey). They measure heat though specialized infrared cameras on their head. They can respond to changes in the thousandth of a degree range. Work in the mid 80's suggested that this sense operates near the limit allowed by the background noise, which hampers signal detection. Since then however, extremely temperature-sensitive ion channels known as TRP receptors has been discovered in thermal sensing neurons. New electrical synchronization phenomena have also been discovered in these neurons. The proposed research will establish, using neural modeling, how important synchronization and TRP receptors are to the thermal sensitivity. This adds a fundamental link in the chain from the infrared optics of the cameras to the higher brain levels that together process infrared and visual images. 2) What is the relation between nerve damage and degradation of information flow? Nerves are a collection of "wires" known as "axons" coming out of single neurons. Fast propagation of information down a nerve, in the form of electrical pulses, relies on myelin. This myelin is made up of specialized cells that wrap around the axon, except at tiny 1 micron gaps called nodes of Ranvier. The myelin modifies the electrical properties of the axon "cable" in a way that pulses quickly jump from node to node. It also insulates one axon from the other. Myelin thins out in diseases like multiple sclerosis (MS). Recently the damage from MS can be quantified at high resolution by state-of-the-art "CARS" microscopy. My colleagues at U. Laval have provided me with images at different stages of the disease in a mouse. Our work proposes to understand and predict functional deficits in these nerves from the images. This involves modeling propagation in healthy and demyelinated single axons and axon bundles. 3) How are brain rhythms altered by stimuli? Debates rage over how rhythms are generated and what they mean. Certain brain rhythms are thought to be important for our ability to pay attention to a stimulus. Our work proposes a simple novel mechanism that can be tested in weakly electric fish brains. Paradoxically, we predict that the rhythm arises in a cell that receives a random signal plus a delayed version of that signal (such "feedforward" circuitry abounds). This will be analyzed in detail using a theory we have recently developed for the activity of cells in networks. This theory will also be expanded to explain how brain rhythms can align themselves to stimulation, a phenomenon missing in certain diseases. The work is theoretical and computational. We formulate predictive mathematical models that are simple enough to explain neural phenomena with a minimal number of biological ingredients. The work will make careful use of existing biological data, and benefit from collaborations with experimentalists. The modeling also serves the purpose of making predictions about function - via theoretical analyses or in silico experiments - which can be tested in new experiments. This research will deepen our knowledge of brain circuit function. It may also lead to novel technologies (e.g. neural prosthetics) based on newly found principles, including technologies to repair faulty circuits or to sense heat.