Fig. 1. SET-seq can be used to profile epigenome and transcriptome in the same sample.

a Schema showing the process of SET-seq. b The Pearson correlations among gene expression libraries constructed from different tagmentation time and temperatures. c The Pearson correlations among gene expression libraries constructed from different numbers of cells. d Heatmaps showing the clustering results of H3K27me3 and H3K4me3 SET-seq. Genes, at which the epigenomic and transcriptional signals were detected, were clustered by hierarchical clustering algorithms using row scaled signal scores. Heatmaps were plotted with column scaled signal scores. e Number of detected genes in scSET-seq and Smart-seq2 in 48 mESCs. Smart-seq2 data were downloaded from GSE151334 [https://www.encodeproject.org/experiments/ENCSR059MBO/] and 48 cells were randomly selected. f Venn diagram showing the overall detected genes between scSET-seq and Smart-seq2 in 48 mESCs. Smart-seq2 data were as in e. g Examples of IGV views of H3K27me3 and H3K4me3 scSET-seq. Signals of histone marks from 200 individual single cells were shown below the signals from aggregated 480 cells. H3K4me3 and H3K27me3 SET-seq results from 10,000 cells (10^4 H3K4me3 and 10^4 H3K27me3) were shown as controls. h The fraction of reads in peaks (FRiPs) of scATAC-seq and scSET-seq. scATAC-seq data were downloaded from GSE100033 [https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE151334]. scSET-seq cells were filtered with a mapping ratio >10%. Paired-Tag and 10x genomics data was same as previous reported⁴³. The boxes were drawn from lower quartile (Q1) to upper quartile (Q3) with the middle line denoting the median, and whiskers with maximum 1.5 IQR (interquartile range). n = 45 (H3K27me3 scSET-seq), 200 (H3K4me3 scSET-seq), 447 (H3K27me3 Paired-Tag), 1,659 (H3K4me3 Paired-Tag), and 93 (scATAC-seq) cells. Source data are provided as a Source Data file.