Single-cell RNA sequencing pseudobulk analysis and machine learning identify candidate biomarkers for ischemic...

Area of Science:

• Cardiovascular Biology
• Molecular Biology
• Genomics

Background:

• Ischemic cardiomyopathy (ICM) results from reduced blood flow to the heart, causing myocardial damage and impaired cardiac function.
• Identifying reliable biomarkers is crucial for understanding ICM mechanisms and developing targeted therapies.

Purpose of the Study:

• To identify novel biomarkers and elucidate regulatory networks associated with ischemic cardiomyopathy (ICM).
• To provide a foundation for further research into ICM pathogenesis and therapeutic interventions.

Main Methods:

• Analysis of single-cell RNA sequencing (scRNA-seq) data to identify cell subpopulations and differentially expressed genes (DEGs).
• Application of machine learning algorithms with Boruta feature selection to identify key disease-characteristic genes from an external ICM dataset.
• Reconstruction of regulatory networks by predicting transcription factors (TFs) and microRNAs (miRNAs).
• Validation of identified biomarkers in an *in vivo* rat model of ICM.

Main Results:

• Four hub genes (MLLT3, GFOD1, COLEC12, RARRES1) were identified as key signatures for ICM, with a combined signature achieving an AUC of 93.0%.
• CTCF was identified as a shared transcription factor regulating these genes.
• Computational analysis predicted 329 miRNAs, with miR-195-5p and miR-5680 showing significant differential expression.
• *In vivo* validation confirmed reduced expression of signature genes and CTCF, and elevated levels of miR-195-5p and miR-5680 in ICM samples.

Conclusions:

• GFOD1, MLLT3, COLEC12, RARRES1, miR-195-5p, and miR-5680 are proposed as promising biomarkers for ischemic cardiomyopathy.
• CTCF is identified as a potential regulatory transcription factor in ICM.
• These findings highlight novel molecular targets for ICM diagnosis and therapy.