A machine learning-derived gene signature of programmed cell death reveals diagnostic biomarkers in dilated cardiomyopathy

Background Dilated cardiomyopathy is a severe myocardial disorder characterized by ventricular dilation and systolic dysfunction. It is associated with poor prognosis due to complex and heterogeneous mechanisms involving immune dysregulation, cell death, and tissue remodeling. Identifying reliable molecular biomarkers and constructing effective diagnostic models remain urgent needs for early detection and targeted therapy. Methods We integrated bulk transcriptomic and single-cell RNA sequencing data from multiple public datasets to explore the molecular landscape of dilated cardiomyopathy. Differential gene expression, weighted gene co-expression network analysis, and machine learning algorithms were used to identify core genes associated with programmed cell death. A diagnostic risk model was established and validated across independent cohorts. Immune infiltration patterns and molecular subtypes were evaluated. Functional pathways and potential therapeutic compounds were explored through gene enrichment analysis, drug–gene interaction databases, and molecular docking. Key gene expression was confirmed by quantitative PCR and Western blot in mouse and human myocardial tissues. Results Eight core genes (AGTR2, GLI2, HRK, IL10, NQO1, NT5E, SFRP1, and STAT4) were identified as significantly altered in dilated cardiomyopathy. These genes demonstrated strong predictive capacity in the risk model across multiple datasets. Immune correlation analysis revealed their association with specific immune cell populations, indicating roles in inflammation and immune remodeling. Consensus clustering revealed distinct molecular subtypes with different immune infiltration profiles. Single-cell analysis showed cell-type–specific expression of core genes in fibroblasts and immune cells. Drug–gene prediction and docking identified several candidate compounds, including resveratrol and folic acid. Experimental validation confirmed consistent expression trends of selected genes at mRNA and protein levels in diseased tissues. Conclusions This study systematically identified key genes associated with programmed cell death in dilated cardiomyopathy and developed a robust diagnostic model. The findings reveal the interplay between immune regulation and cell death mechanisms and highlight novel molecular subtypes and drug targets. These results provide important insights for precise diagnosis and personalized treatment strategies in dilated cardiomyopathy.