Data_Sheet_1_Exploring the key ferroptosis-related gene in the peripheral blood of patients with Alzheimer’s disease and its clinical significance.zip

IntroductionAlzheimer’s disease (AD) is the most common type of dementia, and there is growing evidence suggesting that ferroptosis is involved in its pathogenesis. In this study, we aimed to investigate the key ferroptosis-related genes in AD and identify a novel ferroptosis-related gene diagnosis model for patients with AD.Materials and methodsWe extracted the human blood and hippocampus gene expression data of five datasets (GSE63060, GSE63061, GSE97760, GSE48350, and GSE5281) in the Gene Expression Omnibus database as well as the ferroptosis-related genes from FerrDb. Differentially expressed ferroptosis-related genes were screened by random forest classifier, and were further used to construct a diagnostic model of AD using an artificial neural network. The patterns of immune infiltration in the peripheral immune system of AD were also investigated using the CIBERSORT algorithm.ResultsWe first screened and identified 12 ferroptosis-related genes (ATG3, BNIP3, DDIT3, FH, GABARAPL1, MAPK14, SOCS1, SP1, STAT3, TNFAIP3, UBC, and ULK) via a random forest classifier, which was differentially expressed between the AD and normal control groups. Based on the 12 hub genes, we successfully constructed a satisfactory diagnostic model for differentiating AD patients from normal controls using an artificial neural network and validated its diagnostic efficacy in several external datasets. Further, the key ferroptosis-related genes were found to be strongly correlated to immune cells infiltration in AD.ConclusionWe successfully identified 12 ferroptosis-related genes and established a novel diagnostic model of significant predictive value for AD. These results may help understand the role of ferroptosis in AD pathogenesis and provide promising therapeutic strategies for patients with AD.

阿尔茨海默病（AD）是最常见的痴呆症类型，越来越多的证据表明铁死亡参与了其发病机制。在本研究中，我们旨在探讨阿尔茨海默病中铁死亡相关基因的关键作用，并识别一种针对阿尔茨海默病患者的创新性铁死亡相关基因诊断模型。材料与方法：我们从基因表达综合数据库中提取了五个数据集（GSE63060、GSE63061、GSE97760、GSE48350和GSE5281）中的人类血液和海马基因表达数据，以及来自FerrDb的铁死亡相关基因。通过随机森林分类器筛选差异表达的铁死亡相关基因，并进一步利用人工神经网络构建阿尔茨海默病的诊断模型。此外，我们还利用CIBERSORT算法研究了阿尔茨海默病周围免疫系统中免疫浸润的模式。结果：我们首先通过随机森林分类器筛选并确定了12个铁死亡相关基因（ATG3、BNIP3、DDIT3、FH、GABARAPL1、MAPK14、SOCS1、SP1、STAT3、TNFAIP3、UBC和ULK），这些基因在阿尔茨海默病组和正常对照组之间存在差异表达。基于这12个核心基因，我们成功构建了一个满意的诊断模型，利用人工神经网络区分阿尔茨海默病患者与正常对照组，并在多个外部数据集中验证了其诊断效力。进一步研究发现，关键铁死亡相关基因与阿尔茨海默病中的免疫细胞浸润密切相关。结论：我们成功识别了12个铁死亡相关基因，并建立了一种对阿尔茨海默病具有显著预测价值的创新性诊断模型。这些结果有助于理解铁死亡在阿尔茨海默病发病机制中的作用，并为阿尔茨海默病患者的治疗策略提供了有希望的途径。