Fig. 2: Single-cell ensembles of integrated histone modifications reveal early heterogeneities in 2cell embryos.

a, Schematic of the CoTACIT workflow. PAT-bar 1, PAT assembled with barcoded T5-1 and T7-1 adaptors; PAT-bar 2, PAT assembled with barcoded T5-2 and T7-2 adaptors. b, Co-embeddings of TACIT and CoTACIT data. Each dot represents an individual cell and is coloured by stages and methods. c, Schematic of the pipeline used to generate synthetic cells. d, Top, UMAP visualizations of interpolated 155 single cells based on WNN analysis, which combined all six modalities. Cells are coloured by stages. Bottom, boxplots displaying the average expression of ZGA-related genes for synthetic cells up to the 4cell stage. Genes exhibiting significant upregulation subsequent to the initiation of both minor and major ZGA were categorized as ZGA-related. e, Emission probabilities for each synthetic cell by single-cell ChromHMM. Chromatin-state definitions (left) and genome coverage (right) for each state are annotated. Chromatin-state definitions were determined on the basis of histone-modification probabilities and annotations of genic and non-genic elements (Extended Data Fig. 7a). f, Gene expression associated with chromatin states. Chromatin regions were linked to the nearest genes using Homer. The following number of genomic bins were used: multivalent (Multi), 278; promoter (weak) (Pr-W), 2,920; promoter (strong) (Pr-S), 36,352; enhancer (weak) (En-W), 16,026; enhancer (strong) (En-S), 59,573; gene body (poised) (Ge-P), 12,691; gene body (active) (Ge-A), 20,314; heterochromatin (polycomb) (He-P), 15,246; heterochromatin (H3K9me3) (He-K9), 11,913; and heterochromatin (K27+K9) (He), 23,039. g, Track view displaying chromatin-state annotations in representative loci for synthetic cells. Colours are as for e. For boxplots (d,f), the centre lines indicate the median, box limits indicate the first and third quartiles, and whiskers indicate 1.5× the interquartile range (IQR).

Source data