Abstract:Background Asthma is a heterogeneous respiratory disease with similar clinical picture resulting from different etiologies and pathogenesis. Cuproptosis is a recently discovered novel mode of cell death that is associated with the development of multiple diseases. Our study aims to explore the molecular clusters related to cuproptosis in asthma and to construct clinical prediction models.MethodsCuproptosis regulators and immune signatures in asthma were analyzed based on the GSE41861 dataset. We examined 92 asthma samples, exploring their molecular clusters of cuproptosis-related genes as well as 22 immune cell infiltrations. Differentially expressed genes within each cluster were identified by using the WGCNA algorithm. Then, by evaluating the performance of random forest models, support vector machine models, support vector machine, and eXtreme Gradient Boosting, the optimum machine model was picked. We validated the efficiency of the predictions with column line plots, calibration curves, decision curve analysis and two external datasets.ResultsDifferences between asthmatics and healthy controls in cuproptosis-related genes and immune activation were identified. Two clusters of cuproptosis-related molecules relevant to asthma were identified. Through examining residual values and the area under the curve, the XGB model exhibited promising discriminatory performance. The final XGB model based on 5 genes was constructed and showed better predictive performance in two external validation datasets (AUC = 0.771 and 0.779). Column plots, calibration curves and decision curve analysis demonstrated the accuracy of predicting asthma subtypes.ConclusionOur study illustrates the relationship between cuproptosis and asthma, with proposed predictive models to assess the risk of cuproptosis and asthma subtypes.

Identification and Immunologic Signature of Cuproptosis-Related Molecular Clusters in Asthma