Identification of risk genes associated with myocardial infarction based on the recursive feature elimination...

Area of Science:

• Genomics
• Cardiovascular Research
• Bioinformatics

Background:

• Myocardial infarction (MI) poses a significant global health burden.
• Identifying genetic risk factors is crucial for developing effective diagnostics and therapeutics.
• Gene expression profiling offers a powerful approach to uncover disease-associated genes.

Purpose of the Study:

• To identify novel risk genes associated with myocardial infarction (MI).
• To develop a predictive classifier for MI using identified risk genes.
• To validate the classifier's efficacy on an independent dataset.

Main Methods:

• Differential gene expression analysis of microarray datasets (GSE34198, GSE61144).
• Construction and analysis of a protein-protein interaction (PPI) network.
• Application of the neighboring score method and recursive feature elimination (RFE) algorithm.
• Development and validation of a support vector machine (SVM) classifier.

Main Results:

• Identified 724 downregulated and 483 upregulated differentially expressed genes (DEGs) in MI samples.
• Constructed a PPI network with 1,083 nodes and 46,363 connections.
• Selected 15 risk genes using RFE, achieving 88% precision with an SVM classifier.
• Validated classifier efficacy on GSE61144 with 0.92 predictive precision.
• Highlighted AKAP12 and GLRA2 as significant risk genes.

Conclusions:

• AKAP12 and GLRA2 are identified as potential risk genes in myocardial infarction development.
• These genes may influence cardiac contractility and protect against ischemia-reperfusion injury.
• The findings suggest AKAP12 and GLRA2 as potential diagnostic biomarkers or therapeutic targets for MI.