以往的研究表明，阿尔茨海默病（AD）与许多类型的肿瘤之间存在流行病关联，但连接这些疾病的内在生物学机制目前尚不清楚。在本研究中，我们使用微胶质细胞作为桥梁，从基因和网络水平探索了起源于脑的AD和多形性胶质母细胞瘤（GBM）之间的转录组关联。首先，我们从基因表达清单（GEO）数据库中提取了人类单细胞RNA测序数据集，并在细胞注释后确定了不同表达的基因。观察到11个常见的基因是AD和GBM失调基因所共有的。接下来，我们利用DIAMOnD和Flow Centrality算法根据全局网络拓扑结构识别微胶质细胞模块和介导这两种疾病之间的途径。在这些候选途径中，介导基因FURIN和BACE1（从SPIKN5到CSNK1A1）不仅与AD患者大脑中积累的淀粉样蛋白斑块的形成有关，而且涉及癌症生物学。此外，连接AD和GBM模块的介导途径的生物探索揭示了炎症反应、脂代谢紊乱和细胞增殖等方面。最后，我们根据这些途径涉及的介导基因确定了早期AD检测的新标志物以及神经胶质瘤预后的风险模型。总之，本研究提供了一种新的基于网络的策略来探索AD和GBM之间的微胶质细胞介导，并确定了疾病检测和预后的候选标志。

Previous studies have revealed that there existed epidemic associations between Alzheimer's disease (AD) and many types of tumors, however, the inner biological mechanism connecting these diseases was not clear currently. In this study, we explored the transcriptome associations between AD and glioblastoma multiforme (GBM) that both originate in the brain, using microglia as a bridge, from gene and network levels. Firstly, we extracted human scRNA sequencing datasets from Gene Expression Omnibus (GEO) database, and identified differentially expressed genes within microglia after cell annotation. It was observed that there were 11 common genes shared by AD and GBM dys-regulated genes. Next, we utilized DIAMOnD and Flow Centrality algorithms to identify microglia modules and mediating pathways connecting these two diseases based on global network topology. Among these candidate pathways, the mediating genes FURIN and BACE1 (from SPIKN5 to CSNK1A1) were not only related to the formation of amyloid beta plaques that accumulate in the brain of AD patients, but also involved in cancer biology. Furthermore, the biological explorations of mediating pathways connecting AD and GBM modules reveal inflammatory response, lipid metabolism disorder, and cell proliferation terms. Finally, novel signatures for early AD detection as well as risk models for glioma prognosis were identified based on mediating genes involved in these pathways. In conclusion, this study provided a novel network-based strategy for exploring microglia mediation between AD and GBM and identified candidate signatures for disease detection and prognosis.