Neural Simulations as a Tool in Drug Discovery

描述（由申请人提供）：我们的长期目标是创建足够复杂的计算机大脑模拟，用于药物测试。这一目标与NIH寻求认知、情感或行为病理（包括成瘾和抑郁）的药理学治疗的所有院系相关。计算机模拟药物测试将降低药物开发的成本，缩短开发此类疗法所需的时间。它还将有助于消除对认知、情感和行为产生不良影响的治疗。使用适当的数据集来研究海马体功能，我们已经证明了推动朝着这一长期目标前进的概念。因为基于神经生物学的模拟可以预测行为和认知的某些方面，现在是时候将这种模拟的范围扩展到更多的大脑系统，并使这种模拟与神经活性药物相关。这个应用程序集中在一个特殊的病理，严重，急性创伤后应激障碍（PTSD）。创伤后应激障碍不同于传统的抗抑郁药和选择性血清素再摄取抑制剂（SSRI）治疗的抑郁症。这一应用提出了一种新颖的创伤后应激障碍的综合脑理论，并强调了睡眠的关键性质。创伤后应激障碍不仅与睡眠质量差有关，而且噩梦是严重急性创伤后应激障碍最常见的症状。实验动物模型已经产生了创伤对睡眠和做梦影响的数据；因此，所提出的技术面临的挑战是：(1)建立一个再现这些实验结果的计算模型；(2)预测新的治疗配方，以加速改善睡眠质量，特别是梦睡眠（REM）。计算机模拟将解释神经元水平、突触水平和药物受体相互作用水平上的相互作用。作为我们长期目标的一部分，这样的软件将能够预测：(1)基因剂量的影响；(2)行为体验对睡眠的影响。现有计算模型的扩展包括额外的大脑区域和更多的受体系统。此外，新软件必须连接两种不同风格的大脑建模：一种对应于睡眠的生物化学和药理学，做梦，以及使神经元和突触崩溃为标量相互作用的压力的某些方面，另一种对应于前脑皮质系统使用数万个神经元和数十亿个突触再现上下文和情景学习的模型和模拟。具体来说，由海马区、基底外侧杏仁核和新皮层组成的认知调节系统必须相互连接，并与睡眠和应激系统（包括杏仁核中央（CeA）和位于脑干和下丘脑的几个小核）连接。重要的是，即使我们的创伤后应激障碍理论是错误的，这里提出的软件也可以用于模拟睡眠中断的竞争理论。

DESCRIPTION (provided by applicant): Our long-term goal is to create computational brain simulations sophisticated enough for use in drug testing. This goal is relevant to all houses of NIH seeking pharmacological cures for cognitive, emotional, or behavioral pathologies including addictions and depressions. Drug testing by computer simulations will decrease the cost of drug development and shorten the time required to develop such therapies. It will also help eliminate treatments with undesirable effects on cognition, emotion, and behavior. Using appropriate data sets for studying hippocampal function, we have proof of concept that motivates moving toward this long-term goal. Because neurobiologically-based simulations can predict certain aspects of behavior and cognition, it is time to extend the purview of such simulations to more brain systems and to make such simulations relevant to neuroactive drugs. This application focuses on a particular pathology, severe, acute post-traumatic stress disorder (PTSD). PTSD is distinct from depressions treated by the classic antidepressants and selective serotonin reuptake inhibitors (SSRI's). This application presents a novel, integrated brain theory of PTSD and emphasizes the critical nature of sleep. PTSD not only correlates with poor sleep, but nightmares are the most common symptom in severe, acute PTSD. Experimental animal models already produce data on the effects of trauma on sleep and dreaming; therefore, challenges for the proposed technology are (1) to build a computational model that reproduces such experimental results and (2) to predict novel therapeutic formulations for speeding the improvement of sleep quality, especially dream sleep (REM). The computer simulations will account for interactions at the level of neurons, at the level of synapses, and at the level of drug-receptor interactions. As part of our long-term goal, such software will be able to predict: (1) the effect of gene doses and (2) the effect of behavioral experiences on sleep. Extension of the existing computational models includes additional brain regions and more receptor systems. Moreover, the new software must interface two different styles of brain modeling: one corresponding to the biochemistry and pharmacology of sleep, dreaming, and certain aspects of stress that collapses neurons and synapses into scalar interactions versus the other which corresponds to models and simulations that reproduce contextual and episodic learning by forebrain cortical systems using tens of thousands of neurons and billions of synapses. Specifically, the cognition-mediating system consisting of the hippocampal formation, the basolateral amygdala and neocortex must be interfaced with each other and with the sleep and stress systems, including the n. centralis amygdala (CeA) and several small nuclei located in the brainstem and hypothalamus. Importantly, even if our theory of PTSD is wrong, the software being proposed here has use for modeling competing theories of sleep disruption. Our long-term goal is to create computational brain simulations that are sophisticated enough for drug testing; this goal is relevant to all houses of NIH seeking pharmacological cures for cognitive, emotional, or behavioral pathologies including addictions and depressions. Drug testing by computer simulations will decrease the cost of drug development and shorten the time required to develop such therapies, and it will also help eliminate treatments with undesirable effects on cognition, emotion, and behavior.