Figure 3. Distribution of cardiomyocyte sSNVs with gene-level annotation.
a-b, Distribution of sSNVs across genic regions (a) and across functional categories (b) in IHD and age-matched control. IHD cardiomyocytes harbor a significantly higher burden of sSNVs predicted to be protein-altering (P < 0.05, two-sided Wilcoxon test). c, Natural selection against sSNVs assessed by the exonic-to-intronic ratio and nonsynonymous-to-synonymous ratio (dN/dS), normalized by germline sSNVs. Both control and IHD cardiomyocytes exhibit ratios greater than one, with stronger significance in IHD cardiomyocytes (P = 1.50 × 10−4 and 1.15 × 10−5, two-sided Fisher’s exact test), suggesting relaxed negative selection on somatic mutations compared to germline mutations. d-e, Association between somatic mutations and gene transcription or chromosome accessibility. Solid lines represent the sSNV enrichment ratio with error bars representing the standard deviation, while dashed lines indicate the least-squares regression fit across eight groups classified by their transcription or chromosomal accessibility levels. In both control and IHD cardiomyocytes, sSNVs tend to be enriched in regions with low gene expression and relatively closed chromatin. f-g, GO enrichment analysis of genes harboring deleterious sSNVs in control cardiomyocytes (f) or IHD cardiomyocytes (g). Control cardiomyocytes show an enrichment of deleterious somatic mutations in pathways related to heart muscle development, cytoskeletal organization, and cardiac muscle contraction. In contrast, IHD cardiomyocytes exhibit an enrichment of somatic mutations in pathways associated with cellular response to stress, regulation of mRNA splicing, and acetylation-dependent protein binding.