Development of immune-derived molecular markers for coronary heart disease via multimachine learning

Hao Wang1, Chong Du1, Tongtong Yang1

  • 1Department of Cardiology, the First Affiliated Hospital with Nanjing Medical University, 300 Guangzhou Road, Nanjing 210029, China.

Iscience
|March 5, 2026
PubMed

Insights

Researchers developed a new diagnostic model for coronary artery disease (CAD) using machine learning. The study also identified treprostinil as a potential therapeutic agent to treat CAD, improving endothelial cell function.

Area of Science:

  • Cardiovascular Medicine
  • Genomics
  • Immunology
  • Computational Biology

Background:

  • Coronary artery disease (CAD) poses a significant health burden, with early diagnosis remaining a challenge.
  • Coronary atherosclerosis is the primary cause of CAD.
  • Novel diagnostic and therapeutic strategies are crucial for managing CAD.

Purpose of the Study:

  • To develop a novel, high-performance diagnostic model for CAD using machine learning.
  • To identify potential therapeutic agents for CAD treatment.
  • To investigate the role of immune cell alterations in CAD pathogenesis.

Main Methods:

  • Utilized Gene Expression Omnibus datasets for CAD-associated gene expression analysis.
  • Applied multiple machine learning algorithms to identify CAD-related genes.
  • Conducted immune cell infiltration analysis and in vitro functional assays.

Main Results:

  • Identified 32 key genes associated with CAD and immune responses.
  • Developed a diagnostic model with excellent performance for CAD detection.
  • Observed abnormal alterations in CD8+ T cells and naive B cells in CAD patients.
  • Treprostinil demonstrated protective effects against endothelial cell apoptosis and inflammation.

Conclusions:

  • A robust diagnostic model for CAD was successfully established.
  • Treprostinil shows promise as a therapeutic agent for CAD by mitigating endothelial dysfunction.
  • The findings offer new avenues for CAD diagnosis and treatment strategies.