AUTHOR=Lai Yongxing , Lin Chunjin , Lin Xing , Wu Lijuan , Zhao Yinan , Lin Fan TITLE=Identification and immunological characterization of cuproptosis-related molecular clusters in Alzheimer's disease JOURNAL=Frontiers in Aging Neuroscience VOLUME=Volume 14 - 2022 YEAR=2022 URL=https://www.frontiersin.org/journals/aging-neuroscience/articles/10.3389/fnagi.2022.932676 DOI=10.3389/fnagi.2022.932676 ISSN=1663-4365 ABSTRACT=Introduction: Alzheimer's disease (AD) is the most common dementia with clinical and pathological heterogeneity. Cuprotosis is a recently reported form of cell death, which appears to result in the progression of various diseases. Therefore, the aim of our study was to explore cuprotosis-related molecular clusters in AD and construct a prediction model. Methods: Based on the GSE33000 dataset, we analyzed the cuprotosis regulators expression profiles and immune characteristics in AD. Using 310 AD samples, we explored the molecular clusters based on cuprotosis-related genes (CRGs), along with the related immune cell infiltration. Cluster-specific differentially expressed genes (DEGs) were identified using WGCNA algorithm. Subsequently, the optimal machine model was chosen by comparing the performance of random forest model (RF), support vector machine model (SVM), generalized linear model (GLM), and eXtreme Gradient Boosting (XGB). Nomogram, calibration curve and three external datasets were applied for validate the predictive efficiency. Results: The dysregulated CRGs and activated immune responses were determined between AD and non-AD controls. Two cuprotosis-related molecular clusters were defined in AD. Analysis of immune infiltration suggested the significant heterogeneity of immune between distinct clusters. Cluster2 was characterized by elevated immune score and relatively higher levels of immune infiltration. GSVA analysis showed that cluster-specific DEGs in Cluster2 were closely related to various immune responses. RF machine model presented the best discriminative performance with relatively lower residual and root mean square error (RMSE), and higher area under the curve (AUC=0.9829). A final 5-gene based RF model was constructed, which exhibited satisfactory performance in two external validation datasets (AUC= 0.8529 and 0.8333). The nomogram and calibration curve also demonstrated the accuracy to predict AD subtypes. Further analysis revealed that these five model-relate genes were significantly associated with the Aβ-42 levels and β-secretase activity. Conclusion: Our study systematically illustrated the complicated relationship between cuprotosis and AD, and developed a promising prediction model to evaluate the risk of AD subtype and the pathological outcome of AD patients.