Machine Learning for Chromatin Regulators in Coronary Artery Disease Diagnosis

OBJECTIVES: This study aims to investigate the mechanisms underlying the role of chromatin regulator-related genes (CRRGs) in coronary artery disease (CAD) and develop a diagnostic model for CAD. METHODS: We downloaded CAD datasets from the GEO database and utilized R software for machine learning, modeling, and classification of CAD based on CRRGs. RESULTS: The random forest model was found to be the best approach, identifying USP44, MOCS1, SSRP1, ZNF516, and SCML1 as the top contributing genes for CAD diagnosis and prevention. Differentially expressed CRRGs were associated with aberrant immune cell infiltration in CAD patients. CAD patients were classified into two subtypes based on the expression of differentially expressed CRRGs. The differential expression analysis identified MMP9, LCE1D, LOC92659, SYNGR4, EN2, CACNA1E, GPR78, and LOC92249 as differentially expressed genes distinguishing the two subtypes of CAD. Functional analyses revealed that the differentially expressed genes are enriched in biological processes related to cellular functions, such as responses to metal ions and inorganic substances. The enriched pathways included inflammation and hormone-related pathways, such as IL-17 signaling, endocrine resistance, TNF signaling, and estrogen signaling pathways. CONCLUSION: CAD is associated with CRRGs, which may represent a new direction for CAD treatment.