Unraveling the Genetic Drivers of Heart Failure from Cardiac Endothelial Cells via Single-Cell RNA-Sequencing Data and Machine Learning Model

Background Heart failure is a complex syndrome that hinders the heart's ability to provide oxygen to the tissues and is a significant cause of death globally. Given that left ventricular failure is more common than left atrial failure, this study utilizes single-cell RNA-sequencing data to detect differentially expressed genes (DEGs) between the endothelial cells of the two chambers and examines whether these DEGs are related to heart failure development. Method The analysis of a healthy human dataset was performed using Seurat, an R package, to identify differentially expressed genes (DEGs) between endothelial cells from the left ventricle and the left atrium of the heart. These findings were validated using two datasets that included both humans and mice with and without heart disease. The overlapped DEGs from the datasets were then utilized to develop a risk prediction model by using linear regression, which can determine an individual's risk for heart failure based on the expression levels of the overlapped DEGs. Results Seven genes, including MYL3, MYH6, TNNC1, FN1, B2M, MYL2, and SAT1, were identified with a significant p-value in all three datasets. Enrichment analysis has shown that these genes play a critical role in muscle contraction and heart regulation, and mutations in these genes have been linked to cardiomyopathy. The risk prediction model had a high accuracy rate of 85% in an independent validation dataset. Conclusion This study has successfully identified significant genes in endothelial cells that are associated with heart failure and may explain the difference in morbidity between the left ventricle and left atrium.