A Robust, Automated, Flexible System for Mouse Behavioral Informatics

描述（由申请人提供）： 破坏中枢神经系统（CNS）的疾病是主要的公共卫生问题，对受影响的个人，他们的家庭和社会造成毁灭性的后果。了解中枢神经系统疾病的潜在机制和发现有效的新治疗方法，这些疾病依赖于关键的临床前在体内的行为研究。尽管遗传学和分子生物学取得了巨大的进步，但CNS药物在人体临床试验中的成功率仅为8%。对于许多人类CNS疾病，疾病病理学知之甚少，并影响多个行为领域，导致复杂的表型。这就需要在动物模型中进行全面的行为评估，以捕获在人类CNS疾病中经常观察到的行为中断的广度。然而，目前的自动行为监控方法并没有提供所需的分析工具来分类和整合广泛和连续的行为数据流到一个可解释的背景。已经开发了一种新型的学术原型系统，用于收集小鼠在家庭笼中表现出的多种行为的长时间，高分辨率数据（例如喂食/饮水和休息/活动，昼夜节律夹带的模式）。类似的行为经常被人类疾病过程所破坏。这个学术原型使用复杂的算法来识别基本的行为构建块，并在广泛的时间尺度上表征它们的调节和协调。行为分析的这一创新使该原型成为开发新的临床前行为评估技术的独特平台。该提案的短期目标是将学术原型的基本功能转换为可扩展的商业系统，用于自动化小鼠笼数据收集和分析。这将（1）大大减少行为研究所需的时间和成本，（2）提高检测实验对行为影响的灵敏度，（3）提高数据的可重复性，（4）为综合行为分析提供新的见解，（5）补充或取代标准的个体和单独的局部行为测试。我们的长期目标是开发一种技术，显著提高CNS疾病（如肥胖、成瘾、失眠、神经变性、焦虑和情感障碍）药物治疗的早期可预测性。 公共卫生相关性： 破坏中枢神经系统（CNS）的疾病是主要的公共卫生问题，对受影响的个人，他们的家庭和社会造成毁灭性的后果。CNS药物在人体临床试验中的成功率较低（约8%），可以通过开发更好的动物模型临床前行为评估工具来提高。该提案旨在开发一种独特而强大的新技术，用于对临床前CNS疾病模型进行稳健、自动化和全面的行为分析，以促进发现CNS疾病的新治疗方法。

DESCRIPTION (provided by applicant): Diseases that disrupt the central nervous system (CNS) are major public health problems with devastating consequences for afflicted individuals, their families, and society. Understanding the mechanisms underlying CNS diseases and discovering efficacious new treatments for these disorders relies critically on preclinical in vivo behavioral research. Despite tremendous advances in genetics and molecular biology, the success rate for CNS drugs in human clinical trials is only 8%. For many human CNS disorders, disease pathology is poorly understood and impacts multiple behavioral domains resulting in complex phenotypes. This creates a need for comprehensive behavioral assessment in animal models to capture the breadth of behavioral disruption so frequently observed in human CNS diseases. However, current approaches for automated behavioral monitoring do not provide the analytic tools required to categorize and integrate broad and continuous behavioral data streams into an interpretable context. A novel academic prototype system has been developed to collect long-duration, high-resolution data for multiple behaviors exhibited by mice in home cages (e.g. patterns of feeding/drinking and rest/activity, circadian entrainment). Similar behaviors are frequently disrupted by human disease process. This academic prototype uses sophisticated algorithms to identify fundamental behavioral building blocks and characterize their regulation and coordination over a wide range of time scales. This innovation in behavioral analysis makes this prototype a unique platform for the development of a new preclinical behavioral assessment technology. The short-term goal of this proposal is to convert the basic functionality of the academic prototype into a scalable commercial system for automated mouse home cage data collection and analysis. This will (1) greatly decrease the time and cost required for behavioral investigations, (2) increase the sensitivity with which experimental effects on behavior may be detected, (3) increase reproducibility of data, (4) provide novel insights into integrated behavioral analysis, and (5) complement or replace standard individual and separate focal behavioral tests. Our long-term goal is to develop a technology that significantly improves early predictability of drug therapies for CNS diseases such as obesity, addiction, insomnia, neurodegenerative, anxiety, and affective disorders. PUBLIC HEALTH RELEVANCE: Diseases that disrupt the central nervous system (CNS) are major public health problems with devastating consequences for afflicted individuals, their families, and society. The success rate for CNS drugs in human clinical trials is low (~8%) and could be improved by developing better tools for preclinical behavioral assessment in animal models. This proposal aims to develop a unique and powerful new technology for robust, automated, and comprehensive behavioral analysis of preclinical CNS disease models to facilitate discovery of new treatments for CNS diseases.