Fig. 5
From: EpiCHAOS: a metric to quantify epigenomic heterogeneity in single-cell data

EpiCHAOS is applicable to single-cell epigenomics data from any modality. A Barplots comparing epiCHAOS scores (epiCHAOS) across different lineages of mouse gastrulation using scNMT-seq DNA methylation data from Argelaguet et al. [42] Methylation data summarized across promoters, gene-bodies or CpG islands were used for epiCHAOS computation. Epiblast is colored in blue. B Scatter plot comparing promoter-wide epiCHAOS scores across different gastrulation lineages using single-cell DNA methylation [epiCHAOS (DNAm)] and ATAC-seq [epiCHAOS (ATAC)] data from the same cells as in A. Linear regression line is shown with Pearson correlation coefficient and p-value. C UMAP embedding generated from scTAM-seq DNA methylation data from Scherer et al. [44]. Hematopoietic progenitor state and epiCHAOS scores are annotated. GMP, granulocyte monocyte progenitor; EryP, erythroid progenitor; MEP, myeloid/erythroid progenitor; MPP, multipotent progenitor; HSC, hematopoietic stem cell; MkP, megakaryocyte progenitor. D EpiCHAOS scores calculated using scCHIP-seq data for H3K27me3 from Grosselin et al. [42, 45]. Cells were derived from a PDX breast cancer model, separated based on sensitivity or resistance to capecitabine. Ten subsamples of 100 cells each were taken per condition. Boxplot shows comparison of epiCHAOS scores (epiCHAOS) between sensitive and resistant cells