快感缺失是抑郁症（MDD）的核心症状并显著影响疾病的发生、发展和转归，但其发生机制不详。我们的前期研究发现MDD快感缺失具有独特的灰质网络特征，即扣带回-额叶-眶额叶灰质网络体积越小，则快感缺失症状越严重。本研究将综合运用基于先验知识基础上的单变量分析方法和数据驱动的多变量分析方法，利用高同质性大样本资源优势，探究MDD快感缺失背后的不同模态的特征性脑网络（脑灰质和静息态脑功能网络）；然后纳入与快感缺失密切相关的3条单胺通路（多巴胺通路、5-羟色胺通路和谷氨酸-γ-氨基丁酸通路）的遗传信息，观察通路内和通路间的遗传信息交互和累积效应对脑网络的影响，探寻快感缺失相关脑网络背后的遗传机制。最后利用机器学习的方法结合已获得的稳定的影像遗传学特征建立快感缺失亚型MDD个体化自动诊断模型，并进一步评估其对8周草酸艾司西酞普兰疗效的预测能力，以期为MDD个体化诊治的突破带来希望。

Anhedonia is the core symptom of major depressive disorder (MDD) and significantly affects the occurrence, development and outcome of MDD, but its mechanism is unknown. Our previous study revealed that the anhedonia of MDD had unique gray matter network, that is, the smaller the cingulate-frontal-orbitalfrontal network, the more severe anhedonia. In this study, we combine the univariate analysis based on prior knowledge and the data-driven multivariate analysis to explore the characteristic gray matter network and resting-state brain function network behind the anhedonia of MDD by making use of the advantage of large sample resources with high homogeneity. Then, we focus on the genetic information of 3 monoamine pathways (dopamine pathway, serotonin pathway and glutamate-gamma-aminobutyric acid pathway) closely related to anhedonia to study the influence of the interaction and accumulation of genetic information within and between pathways on the brain network, and to explore the genetic mechanism behind the anhedonia related brain network. Finally, a personalized automated diagnosis model of anhedonia subtype MDD is established by using the method of machine learning based on the obtained stable imaging genetic characteristics. Using the model, we further evaluate its predictive ability on the efficacy of 8-week escitalopram. So as to bring hope for the breakthrough in the personalized diagnosis and treatment of MDD.