Immunophenotyping identifies key immune biomarkers for coronary artery disease through machine learning

Area of Science:

• Immunology
• Cardiovascular Disease Research
• Bioinformatics

Background:

• Coronary artery disease (CAD) immune subtypes, their relationships, and biomarkers are not fully understood.
• Investigating immune cell differences is crucial for understanding CAD pathogenesis.

Purpose of the Study:

• To characterize immune subtypes in CAD.
• To identify immune biomarkers and key genes associated with CAD.
• To develop predictive models for CAD.

Main Methods:

• Utilized public datasets (GSE20686, GSE42148) for analysis.
• Employed Principal Component Analysis (PCA) and Gene Set Variation Analysis (GSVA).
• Conducted Gene Ontology (GO), Kyoto Encyclopedia of Genes and Genomes (KEGG), and machine learning analyses for prediction and validation.

Main Results:

• Identified two distinct immune cell subtypes with significant differences in programmed cell death-ligand 1 (PD-L1) expression.
• Discovered three CAD hub genes via Weighted Gene Co-expression Network Analysis (WGCNA).
• Developed a Random Forest (RF) model using AKT1 and PTK2B, showing their differential expression in cardiac myocytes and PTK2B's role in cell proliferation and apoptosis.

Conclusions:

• Immunophenotyping revealed a link between CAD and PD-L1.
• AKT1 and PTK2B are identified as key signature genes in CAD.
• These genes may offer clinical utility for CAD diagnosis, prognosis, and treatment.