Atrial fibrillation risk model based on LASSO and SVM algorithms and immune infiltration of key mitochondrial energy metabolism genes

Atrial fibrillation (AF) is a predominant cardiac arrhythmia with unclear etiology. This study used bioinformatics and machine learning to explore the relationship between mitochondrial energy metabolism-related genes (MEMRGs) and immune infiltration in AF. The datasets GSE31821, GSE41177, and GSE79768 were retrieved from the Gene Expression Omnibus (GEO) database, and differential expression analysis identified 59 mitochondrial energy metabolism-related differentially expressed genes (MEMRDEGs) associated with AF. Key MEMRDEGs were selected using the least absolute shrinkage and selection operator (LASSO) and support vector machine (SVM) methods, and a diagnostic model was developed. Immune infiltration was assessed using single-sample gene set enrichment analysis (ssGSEA) and the microenvironment cell population counter (MCPcounter). The diagnostic model, based on the key genes ACAT1, ALDH1L2, HTT, OGDH, and SLC25A3, achieved an area under the curve (AUC) of 0.903. Significant differences in immune cell composition were observed between the AF and control groups. ALDH1L2 was positively correlated with most immune cells, while SLC25A3 showed a negatively correlated with the monocytic lineage. The findings indicate that MEMRGs interact with immune responses in AF, offering insights into the potential molecular mechanisms and therapeutic targets for AF.