Unraveling the cellular characteristics of cardiomyopathy with rare variant-driven gene signatures using multi-omics...

Whole-genome sequencing technologies have advanced, leading to an increase in uncharacterized variants with unknown functions. This study focuses on characterizing cardiomyopathy-associated genes harboring theses rare variants and uncovering their cellular contexts using single-cell transcriptomics. We investigated whole genome sequencing on 245 unrelated Korean patients with either dilated (48.2%) and hypertrophic (47.8%) cardiomyopathy. Rare variants were identified and subjected to burden analysis to detect potentially causative candidates. To understand their biological impact, we profiled single-cell transcriptomes from an independent dataset, highlighting cell populations and interactions relevant to disease pathogenesis. A total of 3584 rare variants were discovered, including 50 pathogenic or likely pathogenic variants in 41 patients. Among the remaining 3534 variants of uncertain significance (VUS), burden analysis revealed 144 gene signatures significantly enriched in pathways related to cardiac muscle tissue development, heart morphogenesis, and endocrine system development (FDR < 0.003). These gene signatures were strongly correlated with cardiomyopathy - associated phenotypes, including HCM (HP:0001639) and DCM (HP:0001644) from the Human Phenotype Ontology (HPO) database. Single-cell transcriptomic analysis of 11,664 heart tissue cells demonstrated that the expression of these gene is influenced by cellular heterogeneity, which may contribute to disease manifestation. Further analysis identified five major cell types in heart tissue, revealing dynamic interactions between cardiomyocytes and endothelial cells. Rare genetic variants, including VUS, are closely linked to cardiac developmental processes and the sarcolemma integrity in cardiomyopathy. Our findings underscore the critical involvement of both cardiomyocytes and non-cardiomyocytes in disease etiology, highlighting the utility of integrating genomic and single-cell data for mechanistic insights.