Characterising Shared and Specific Cell–Cell Communication in Cardiomyopathy Subtypes From Single‐Cell Transcriptomics Data

Figures

• 

  FIGURE 1. Identification of Major Cell Compartments in Human Cardiomyopathy and Healthy Donors through Single‐cell Data Integration. (A) Overview of Dataset Integration, Downstream Analysis and Interpretation of Results Using Single‐Cell RNA Sequencing Datasets from Three Independent Studies. (B) UMAP embeddings of ‘celltype.l2’ cell types, split by disease subtype and healthy donor group. Cell types are labelled by abbreviation as follows: Adipocyte (AD), Arterial Endothelial cells (aECs), Atrial Cardiomyocyte cells (aCMs), B cells (B), Capillary Endothelial cells (cECs), Endocardial (Endo), Endothelial cells (ECs), Fibroblasts (FBs), Innate Lymphoid cells (ILCs), Lymphatic Endothelial cells (lECs), Macrophages (MPs), Mast (Mast), Mesothelial cells (MESO), Monocyte/cDC (Mono/cDC), Neuronal cells (N), NK/T cells (NK), Pericytes (Per), Smooth Muscle cells (SMCs), T cells (T), Venous Endothelial cells (vECs) and Ventricular Cardiomyocyte cells (vCMs). (C) Stack violin plots showing the ‘celltype.l2’ annotated cell types specific expression of the top‐ranking candidate marker genes for each identified cell type across disease subtype and healthy donor group.

• 

  FIGURE 2. Characterise Cell Composition in Each Patient Subgroup and Evaluate Cell Type‐Specific Disease Preferences. (A) Stack bar plots showing the distribution of Level_2 annotated cell types across disease subtypes and the healthy group. (B) Heatmap showing disease preference for each annotated cell type, evaluated by the ratio of observed to expected cell counts (Ro/e) value. (C) UMAP plot showing the expression patterns of the ILC marker gene KIT in annotated ILC cells across detected patient groups. (D) UMAP plot showing the expression patterns of the genes, including LY96(MD‐2), CPA3, NRXN1, RYR2, LINC00639 and ITGA8 in annotated Monocyte/cDC, Mast, Atrial Cardiomyocyte, Neuronal, Arterial Endothelial and Smooth Muscle, respectively. Split by disease subtype and healthy donor group. Dot colours represent expression level.

• 

  FIGURE 3. Gene Set Enrichment Analysis Reveals Hallmark Gene Sets Across Cardiomyopathy Subtypes and Annotated Cell Types. (A) Dot plot showing Hallmark gene‐set enrichment analysis comparing disease versus healthy controls across four cardiomyopathy subtypes (ACM, DCM, HCM, ICM). The plot is faceted by disease subtype, with the x‐axis representing the normalised enrichment score (NES) and the y‐axis indicating pathway names. Dot size corresponds to the −log10 (p‐value), and dot colour indicates statistical significance. (B) Violin plot showing single‐sample gene set enrichment analysis (ssGSEA) scores for pathways identified in Panel A across annotated cell types. The plot is faceted by enriched pathways from each disease subtype, with the x‐axis and violin colour representing the annotated cell types and the y‐axis indicating the ssGSEA score.

• 

  FIGURE 4. Summary of cell–cell communication in patients. Interactome analysis was performed on the annotated scRNAseq analysis of disease vs. healthy control using the MultiNicheNet package. Circos plots showing the top 30 ligand‐receptor interactions differentially activated in disease subtypes compared to healthy controls: ACM (A), DCM (B), HCM (C) and ICM (D). All clusters listed in the sender and receiver columns within the respective disease group of the prioritised table were included.

• 

  FIGURE 5. Summary of Shared and Disease‐Specific Cell–Cell Communications Across Disease Subtypes. Bubble plots display ligand‐receptor pairs and their differential impact across disease subtypes: (A) shared across three or more disease subtypes, (B) preferentially affected in ACM, (C) preferentially affected in DCM, (D) preferentially affected in HCM and (E) preferentially affected in ICM. Each plot includes scaled ligand activity in receiver cells, scaled cell type activity and curation levels of intercellular communication based on the Omnipath dataset.

• 

  FIGURE 6. Validation of Disease‐Enriched Pathways in Cardiomyopathy Mouse Models Using ssGSEA. Heatmaps showing single‐sample gene set enrichment analysis (ssGSEA) scores for hallmark gene sets in mouse models of cardiomyopathy. Each panel represents one disease subtype: (A) ACM, (B) DCM, (C) HCM and (D) ICM. Rows correspond to annotated hallmark gene sets (left), and columns represent individual samples (top), grouped by disease versus wild type (WT). The colour scale ranges from blue (low ssGSEA score) to red (high ssGSEA score), reflecting pathway activity. This analysis validates the pathways identified through GSEA in human cardiomyopathy by assessing their enrichment in corresponding mouse models.