本研究综合应用神经影像学和转录组学等前沿技术，系统地探究了重性抑郁症的神经机制，并特别关注正常人群中抑郁与焦虑的共病症状，运用结构方程模型等方法揭示其神经基础。在第一项研究中，我们重点探讨了miR-134与MDD之间的联系，采用并行的两阶段跨尺度多组学方法，结合微观分子生物学与宏观神经影像学，深入理解MDD的复杂病理机制。通过对两个不同队列使用磁共振成像得到的数据进行分析，我们揭示了MDD患者中前扣带皮层（ACC）的功能性连接障碍模式，并通过群体分析建立了从miR-134靶基因的分子组织到宏观大脑功能性连接障碍及相关行为的层级跨尺度关联。此外，我们通过转录组学技术发现，外泌体中miR-134的表达水平对MDD临床症状的个体差异，特别是通过调节ACC的功能性连接障碍来影响症状表现。这一发现不仅揭示了miR-134与抑郁之间在基因-大脑-行为层面的多尺度相关性，而且突显了其在调节前扣带皮层功能连接障碍中的关键作用，为miR-134作为MDD治疗潜在生物标志物的重要性提供了新视角。在第二项研究中，我们利用UK Biobank的大样本数据，采用症状级建模方法探究抑郁与焦虑共病症状的高阶共享因子。分析显示，抑郁和焦虑共病症状与特定脑区灰质体积和白质纤维束完整性的异常相关联，这些区域包括胼胝体、尾状核、中央前回等灰质区域，以及小脑上脚、海马束、小脑下脚等白质区域。研究的发现为抑郁与焦虑共病症状的神经基础提供了新的见解。

This study systematically investigates the neural mechanisms of Major Depressive Disorder using cutting-edge techniques in neuroimaging and transcriptomics, with a particular focus on the comorbid symptoms of depression and anxiety in normal people, revealing their neural underpinnings through methods such as structural equation modeling. In the first study, we examined the relationship between miR-134 and MDD, employing an innovative parallel two-stage cross-scale multi-omics approach. This method integrates microscopic molecular biology with macroscopic neuroimaging, facilitating a comprehensive understanding of the complex pathology of MDD. Analyzing data obtained from magnetic resonance imaging in two distinct cohorts, we identified a consistent pattern of functional dysconnectivity in the anterior cingulate cortex (ACC) among MDD patients. Furthermore, through population-based analyses, we established a hierarchical cross-scale association from the molecular organization of miR-134 target genes to macroscopic brain functional dysconnectivity and related behaviors. Additionally, our transcriptomic studies revealed that individual variations in exosomal miR-134 expression levels influence the clinical symptoms of MDD, particularly by modulating ACC-related functional connectivity disruptions. These findings not only demonstrate the multi-scale correlation between miR-134 and depression across gene-brain-behavior dimensions but also highlight its crucial role in regulating ACC functional dysconnectivity, offering new perspectives on the importance of miR-134 as a potential biomarker for MDD treatment. In the second study, utilizing the large-scale data from the UK Biobank, we employed symptom-level modeling to explore the higher-order shared factors of comorbid symptoms of depression and anxiety. Our analysis revealed associations between these comorbid symptoms and abnormalities in gray matter volume and white matter integrity in specific brain regions, including the corpus callosum, caudate nucleus, and anterior cingulate, as well as white matter areas such as the superior cerebellar peduncle, hippocampal tract, and inferior cerebellar peduncle. These insights provide a new understanding of the neuroanatomical basis for depression and anxiety comorbidity.