فهرست نشریات علمی دسترسی آزاد ایران

Background and Objectives: Severe asthma (SA) is a complicated disorder, and its pathogenesis involves various contributing factors. This study is to identify and validate the hub genes associated with SA, in order to determine new biomarkers and therapeutic targets.Methods: We conducted weighted gene co-expression network analysis (WGCNA) using the GSE63142 dataset to identify key modules and genes that might be associated with SA. Furthermore, we used LASSO regression and random forest to determine the hub genes, and constructed a nomogram prediction model for the risk of SA based on the hub genes. Finally, we verified the predictive value of hub genes using the GSE43696.Results: We identified the gray module, and the genes of the gray module are mainly related to Asthma and cytokine-cytokine receptor interactions. This article screened out 16 hub genes from the grey module, including FCER1A, HLA-DPB1, MAOB, FKBP5, TPRXL, SYT8, SLCO1B3, IL20RB, KCNA1, OXTR, SCGB1A1, PROS1, TCN1, C7orf26, IL1R2, SEC14L3. Taking 16 hub genes as predictors, a nomogram model of the SA occurrence risk was constructed. ROC curve analysis revealed that the model demonstrated a predictive specificity of 0.838, the sensitivity was 0.875, and AUC was 0.911 (0.866-0.955). The model's diagnostic effect was further verified using the GSE43696 dataset. The AUC in the validation set was 0.901 (0.846-0.956), the specificity and sensitivity were 0.771 and 0.921.Conclusion: The SA prediction model constructed based on these 16 hub genes has superior predictive efficacy and can be used as a powerful tool for clinical diagnosis.