Atherosclerotic plaque vulnerability quantification system for clinical and biological interpretability.

Area of Science:

• Computational systems biology
• Biomarker discovery
• Cardiovascular disease research

Background:

• Acute myocardial infarction is a leading cause of death from coronary artery disease.
• Identifying biomarkers for plaque destabilization is crucial for preventing thrombotic events.
• Current diagnostic methods require improvement for assessing atherosclerosis severity.

Purpose of the Study:

• To develop a computational tool for identifying atherosclerotic plaque vulnerability signatures (APVS).
• To create a robust classifier for distinguishing ST-elevation myocardial infarction from chronic coronary syndrome.
• To establish a system for quantifying plaque vulnerability and assessing atherosclerosis severity.

Main Methods:

• Utilized a machine-learning integrative program with nine learners.
• Analyzed 2,235 samples from bulk cohorts and 14 samples from single-cell cohorts.
• Developed an APVS-based quantification system (APVSLevel).

Main Results:

• Generated a reliable warning classifier for atherosclerotic plaque vulnerability.
• Achieved high accuracy in distinguishing ST-elevation myocardial infarction from chronic coronary syndrome.
• Identified 33 pathogenic biomarkers and unraveled multidimensional dysregulated mechanisms.

Conclusions:

• The developed APVS classifier and APVSLevel system offer potential for early diagnosis and severity assessment of atherosclerosis.
• This tool can aid in differential diagnosis and evaluating subtle genetic pathological changes.
• The study provides a novel approach for understanding plaque vulnerability in coronary artery disease.