Probing networks underlying sleep and stress with multiscale data

描述（由申请人提供）：临床和实验文献迅速融合，证明睡眠和昼夜节律障碍在广泛的慢性疾病的发病和维持中起着不可或缺的作用。睡眠障碍在神经精神疾病患者中尤为常见，这是美国疾病负担的主要来源，是心血管疾病的两倍多。流行病学数据表明，睡眠中断在人们焦虑、抑郁和创伤后应激障碍之前就会出现，而且往往会使人们容易出现这种情况，这表明这些复杂的特征是高度交织在一起的。解决睡眠和压力表型的复杂性需要利用和整合多种形式数据的新方法。我的建议侧重于确定睡眠和压力特征共同的因果分子网络，以研究与神经精神疾病相关的新疾病机制和治疗策略。通过详尽的表型分析（479个睡眠和压力特征）和集成的多尺度计算方法，我们将利用这些特征的复杂性来探索分子途径如何作为一个协调单位自然相互作用，而不是当它们被人为操纵时如何反应。通过整合来自F2个小鼠群体（bbb100只小鼠）的遗传、基因表达和共表达数据，我们可以使用贝叶斯重建来识别作为应激和睡眠表型因果调节因子的分子亚网络。由于睡眠和压力特征的复杂性，我们必须在多基因网络的背景下理解个体基因，并将表型视为这些网络的突现特性。通过了解这些分子网络如何作为表型的传感器和驱动因素，我们可以适当地考虑药理学干预的目标，并利用新的计算策略来重新利用药物。我们期望这项工作将为未来与抑郁、焦虑、应激易感性（PTSD）相关的体内研究提供基础。

DESCRIPTION (provided by applicant): Clinical and experimental literature have rapidly converged to demonstrate that sleep and circadian dysfunction play an integral role in the onset and maintenance of a broad spectrum of chronic diseases. Sleep dysfunction is especially common among patients with neuropsychiatric disorders, which is the leading contributor of disease burden in the United States, more than twice as much as cardiovascular disease. Epidemiological data demonstrates that sleep disruption precedes and often predisposes people to anxiety, depression, and PTSD, suggesting that these complex traits are highly intertwined. Addressing the complexity of sleep and stress phenotypes requires novel approaches that leverage and integrate multiple forms of data. My proposal focuses on identifying causal molecular networks common to sleep and stress traits in order to investigate novel disease mechanisms and therapeutic strategies relevant to neuropsychiatric disorders. With an exhaustive phenotypic assay (479 sleep and stress traits) and an integrated multi-scale computational approach, we will leverage the complexity of these traits to probe how molecular pathways naturally interact as a coordinated unit, rather than how they react when they are artificially manipulated. By integrating genetic, gene expression, and co-expression data from F2 mouse populations (>100 mice), we can use Bayesian reconstruction to identify molecular subnetworks that act as causal regulators of stress and sleep phenotypes. Because of the complexity of sleep and stress traits, it is imperative that we understand individual genes in the context of polygenic networks and treat phenotypes as emergent properties of these networks. By understanding how these molecular networks act as sensors and drivers of phenotypes, we can then appropriately consider targets for pharmacological interventions and utilize novel computational strategies for repurposing drugs. We anticipate that this work will provide the foundation for future in vivo studies related to depression, anxiety, stress susceptibility (PTSD), and sleep.