Thrombomodulin as a potential diagnostic marker of acute myocardial infarction and correlation with immune infiltration: Comprehensive analysis based on multiple machine learning

Guoqing Liu,Lixia Huang,Xiangwen Lv,Yuting Guan,Lang Li

DOI: https://doi.org/10.1016/j.trim.2024.102070

Abstract:Background: Acute myocardial infarction (AMI) is a global health problem with high mortality. Early diagnosis can prevent the development of AMI and provide valuable information for subsequent treatment. Angiogenesis has been shown to be a critical factor in the development of infarction and targeting this process may be a potential protective strategy for preventing myocardial injury and improving the prognosis of AMI patients. This study aimed to screen and verify diagnostic markers related to angiogenesis in AMI and to investigate the molecular mechanisms of action associated with AMI in terms of immune cell infiltration. Methods: The GSE66360 and the GSE60993 datasets were both downloaded from the GEO database and were used as the training cohort and the external validation cohort, respectively. Angiogenesis-related genes (ARGs) were downloaded from the MSigDB database. The hub ARGs were identified via LASSO, RF, and SVM-RFE algorithms. ROC curves were used to assess the accuracy of the hub ARGs. The potential mechanisms of the hub ARGs were analyzed by GSEA. The ssGSEA algorithm was used to determine differences in immune cell infiltration and immune function. The CIBERSORT algorithm was used for immune cell infiltration analysis. In addition, we constructed a ceRNA network map of differentially expressed ARGs. Results: We identified the thrombomodulin (THBD) gene from ARGs as a potential diagnostic marker for AMI based on the LASSO, SVM-RFE, and RF algorithms. THBD was differentially expressed and had a potential diagnostic value (area under the curve [AUC] = 0.931 and 0.765 in the training and testing datasets, respectively). GSEA showed that the MAPK signaling pathway was more enriched in the high-expression group of THBD (P < 0.05). Immune cell infiltration analysis demonstrated that THBD was mainly positively correlated with monocytes (R = 0.48, P = 0.00055) and neutrophils (R = 0.36, P = 0.013). Finally, in the ceRNA regulatory network, THBD was closely associated with 9 miRNAs and 42 lncRNAs involved in AMI. Conclusion: THBD can be used as a potential diagnostic marker for AMI. This study provides new insights for future AMI diagnosis and molecular mechanism research. Moreover, immune cell infiltration plays an essential role in the occurrence and development of AMI.