Genomic structural variations lead to dysregulation of important coding and non‐coding RNA species in dilated cardiomyopathy

The transcriptome needs to be tightly regulated by mechanisms that include transcription factors, enhancers, and repressors as well as non‐coding RNAs. Besides this dynamic regulation, a large part of phenotypic variability of eukaryotes is expressed through changes in gene transcription caused by genetic variation. In this study, we evaluate genome‐wide structural genomic variants (SVs) and their association with gene expression in the human heart. We detected 3,898 individual SVs affecting all classes of gene transcripts (e.g., mRNA, miRNA, lncRNA) and regulatory genomic regions (e.g., enhancer or TFBS). In a cohort of patients ( n = 50) with dilated cardiomyopathy (DCM), 80,635 non‐protein‐coding elements of the genome are deleted or duplicated by SVs, containing 3,758 long non‐coding RNAs and 1,756 protein‐coding transcripts. 65.3% of the SV‐eQTLs do not harbor a significant SNV‐eQTL, and for the regions with both classes of association, we find similar effect sizes. In case of deleted protein‐coding exons, we find downregulation of the associated transcripts, duplication events, however, do not show significant changes over all events. In summary, we are first to describe the genomic variability associated with SVs in heart failure due to DCM and dissect their impact on the transcriptome. Overall, SVs explain up to 7.5% of the variation of cardiac gene expression, underlining the importance to study human myocardial gene expression in the context of the individual genome. This has immediate implications for studies on basic mechanisms of cardiac maladaptation, biomarkers, and (gene) therapeutic studies alike. Synopsis In a multi‐omics analysis, structural variants in dilated cardiomyopathy (DCM) patients have been detected that are associated with altered gene expression. Besides regulatory genomic elements, also non‐coding RNAs have been identified as possible new players in the molecular homeostasis in the heart. Germline structural variants in DCM patients are distributed evenly over the genome, with 3.7 times more deletions than insertions. Cis regulatory effects of structural variants can be linked to distinct protein coding and non‐coding transcripts. Cardiac SV‐eQTLs are mainly heart specific compared to peripheral blood and may explain up to 7.5% of transcript variation. Whole genome sequencing seems indispensable when studying the human cardiac transcriptome to avoid larger bias due to otherwise undetected SVs. Graphical Abstract In a multi‐