Integrating circadian, neuroimaging and genetic data to investigate major depression and bipolar disorder

学生战略优先领域：基础和临床研究关键词：昼夜节律、抑郁症、双相情感障碍、遗传学、脑成像。日常昼夜节律的紊乱与情绪障碍和认知能力受损的风险更大相关。然而，到目前为止，大多数研究都使用了节奏性的主观测量，小样本或横截面样本，而没有检查潜在的影响中介和调节因素。这个跨学科的项目将使用具有里程碑意义的英国生物库队列中的时间生物学、神经成像和基因数据。该队列包括50多万人的生活方式、社会人口学和基因数据，以及30,000多人的大脑核磁共振数据；我们的团队已经为100,000人得出了基于客观加速度量学的昼夜休息-活动节奏测量方法。使用这些丰富的数据，机器学习和回归方法可以帮助确定哪些客观节律性变量最能预测情绪障碍和相关结果；以及昼夜节律紊乱对情绪和认知的影响是部分由于对大脑结构的影响，还是受到遗传因素的调节。使用来自英国生物库参与者相关健康记录的数据，将开发模型，以评估是否可以根据昼夜节律、神经成像、遗传、社会人口和生活方式因素的组合来预测情绪障碍发作。该项目将提供与精确医学相关的广泛研究技能方面的培训，包括流行病学、基因组学、神经成像和机器学习。

Studentship strategic priority area:Basic and Clinical ResearchKeywords: Circadian rhythms, depression, bipolar disorder, genetics, brain imaging.Disruption to daily circadian rhythms is associated with greater risk of mood disorder and impaired cognitive ability. However, so far most studies have used subjective measures of rhythmicity, small or cross-sectional samples, and haven't examined potential mediators and moderators of effects. This interdisciplinary project will use chronobiological, neuroimaging and genetic data from the landmark UK Biobank cohort. The cohort includes lifestyle, sociodemographic and genetic data for over 500,000 individuals and brain MRI for over 30,000; and our group has already derived objective accelerometry-based measures of circadian rest-activity rhythms for 100,000 individuals. Using this wealth of data, machine learning and regression methods can help determine which objective rhythmicity variables best predict mood disorder and related outcomes; and whether the influence of circadian disruption on mood and cognition is partly due to effects on brain structure, or is moderated by genetic factors. Using data from linked health records of UK Biobank participants, models will be developed to assess whether mood disorder episodes can be predicted from a combination of circadian rhythmicity, neuroimaging, genetic, sociodemographic and lifestyle factors. This project will provide training in a wide range of research skills relevant to precision medicine, including epidemiology, genomics, neuroimaging and machine learning.