Abstract
OBJECTIVE: This study aimed to identify early biomarkers associated with the progression from atherosclerosis (AS) to heart failure (HF) by integrating single-cell RNA sequencing (scRNA-seq) and bulk transcriptomic data, and to explore the potential underlying mechanisms. METHOD: Transcriptomic datasets (GSE28829 and GSE57345) were obtained from the Gene Expression Omnibus (GEO) database, and single-cell RNA sequencing (scRNA-seq) data were downloaded from the Human Cell Landscape (HCL) platform. Genes of interest were identified by integrating results from weighted gene co-expression network analysis (WGCNA), differentially expressed genes (DEGs) analysis, and cell-type-specific expression patterns. Three machine learning algorithms (LASSO, Random Forest, and SVM-RFE) were employed to screen for robust candidate biomarkers. External validation was performed using three independent datasets: GSE53274, GSE5406, and GSE59867. RESULT: ScRNA-seq data screened for 2828 cardiac-related genes. WGCNA identified 918 genes highly associated with AS. In addition, the limma package identified 9675 DEGs associated with HF progression. A total of 119 overlapping genes were obtained by intersecting the results from the above three analyses. Based on these 119 overlapping genes, three machine learning algorithms (LASSO, Random Forest, and SVM-RFE) were applied to datasets GSE28829 and GSE57345, and consistently identified CD48 as a robust signature gene, with an area under the curve (AUC) greater than 0.7. External validation confirmed CD48 as a potential biomarker for the progression from AS to HF. CONCLUSION: CD48 was identified as a potential early biomarker for the transition from AS to HF, which may offer new insights for risk stratification and early intervention in disease progression.