Identification of key biomarkers for myocardial infarction by multi-omics analysis and machine learning

Background: Acute myocardial infarction (AMI) is one of the leading causes of mortality worldwide. Despite extensive research, only a limited number of genes have been identified as reliable biomarkers for the diagnosis and treatment of AMI. This study aims to identify novel biomarkers and therapeutic targets for AMI by integrating multi-omics data and machine learning. Methods: We obtained the GWAS dataset I9_MI_STRICT from the FinnGen database and the eQTL dataset of peripheral blood from the GTEx database. Using these datasets, we identified genes significantly associated with AMI through transcriptome-wide association studies (TWAS). Functional enrichment analysis was performed using Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathways. Additionally, we downloaded three AMI peripheral blood gene expression microarray datasets (GSE66360, GSE48060, GSE60993) from the Gene Expression Omnibus (GEO) database. Key genes were further identified by combining the risk prediction model constructed by 12 machine learning methods(dataset GSE66360 as the training set, dataset GSE48060 and dataset GSE60993 as the validation set) and Bayesian colocalization analysis. To explore the potential mechanisms of these key genes in AMI, we conducted immunoinfiltration analysis, single-gene Gene Set Enrichment Analysis (GSEA), and Gene Set Variation Analysis (GSVA). Finally, the expression of key genes was validated using real-time quantitative PCR (RT-qPCR) and western blot. Results: showed downregulation compared to the control group. Conclusions: This study identified nine key genes as potential novel targets for the diagnosis and treatment of AMI.