AUTHOR=Zhang Qunhui , Guo Yang , Zhang Benyin , Liu Hairui , Peng Yanfeng , Wang Di , Zhang Dejun TITLE=Identification of hub biomarkers of myocardial infarction by single-cell sequencing, bioinformatics, and machine learning JOURNAL=Frontiers in Cardiovascular Medicine VOLUME=Volume 9 - 2022 YEAR=2022 URL=https://www.frontiersin.org/journals/cardiovascular-medicine/articles/10.3389/fcvm.2022.939972 DOI=10.3389/fcvm.2022.939972 ISSN=2297-055X ABSTRACT=Background: Myocardial infarction (MI) is one of the first cardiovascular endangering human health. Inflammatory response plays a significant role in the pathophysiological process of MI. The object of this study is to screen the expression of mRNA, investigate the function of mRNA, and provide an underlying scientific basis for the diagnosis and treatment of MI. Methods: Four RNA microarray datasets of MI were downloaded from the GSE66360, GSE97320, GSE60993 and GSE48060. The function analysis was carried out by Gene Ontology (GO), Kyoto Encyclopedia of Genes and Genomes (KEGG), and Disease Ontology (DO) enrichment analysis. At the same time, inflammation-related genes (IRGs) were acquired from the GeneCards database. The 52 co-DEGs were acquired from differentially expressed genes (DEGs) in differential analysis, IRGs and genes from SCS, and used to construct a PPI network. Then, three machine learning algorithms, including random forest, least absolute shrinkage and selection operator, support vector machine recursive feature elimination, were used to filter the co-DEGs. Gene set enrichment analysis (GSEA) was performed to screen the hub modulating signaling pathways associated with the hub genes. And these results were validatedin the GSE97320, GSE60993 and GSE48060. The CIBERSORT algorithm was performed to analyze 22 infiltrating immune cells in MI and health controls (CON), and analyze the correlation between these immune cells. Pymol was performed for molecular docking of hub DEGs and potential MI drugs acquired from the COREMINE. Results: A total of 126 DEGs were in the MI and CON groups. After the screening of two machine learning algorithms and key co-DEGs from a PPI network, two hub DEGs (IL1B and TLR2) was obtained. The diagnostic efficiency of IL1B, TLR2, and IL1B+TLR2 showed a good discrimination in four cohort. Immune analysis indicated that IL1B and TLR2 were correlated with various of immune cells. Conclusion: This study identified one hub DEG (IL1B) and illustrated their potential roles in the diagnosis of MI to enhance our knowledge of the underlying molecular mechanism. Infiltrating immune cells played an important role in myocardial infarction. TCM, especially HF, was a potential drug for the treatment of MI.