Multiplex single cell profiling of chromatin accessibility by combinatorial cellular indexing - PubMed (original) (raw)

Multiplex single cell profiling of chromatin accessibility by combinatorial cellular indexing

Darren A Cusanovich et al. Science. 2015.

Abstract

Technical advances have enabled the collection of genome and transcriptome data sets with single-cell resolution. However, single-cell characterization of the epigenome has remained challenging. Furthermore, because cells must be physically separated before biochemical processing, conventional single-cell preparatory methods scale linearly. We applied combinatorial cellular indexing to measure chromatin accessibility in thousands of single cells per assay, circumventing the need for compartmentalization of individual cells. We report chromatin accessibility profiles from more than 15,000 single cells and use these data to cluster cells on the basis of chromatin accessibility landscapes. We identify modules of coordinately regulated chromatin accessibility at the level of single cells both between and within cell types, with a scalable method that may accelerate progress toward a human cell atlas.

Copyright © 2015, American Association for the Advancement of Science.

PubMed Disclaimer

Figures

Fig. 1

Fig. 1. Schematic of combinatorial cellular indexing and validation for measuring single-cell chromatin accessibility

(A) Nuclei are isolated and molecularly tagged in bulk with barcoded Tn5 transposases in wells (labels A–D). Nuclei are then pooled and a limited number redistributed into a second set of wells. A second barcode (labels 1–4) is introduced during PCR. (B) Scatterplot of number of reads mapping uniquely to human or mouse genome for individual barcode combinations. (C) Fragment size distribution for single-cell ATAC-seq vs. published bulk ATAC-seq (4). (D) Boxplot of the fraction of reads mapping to ENCODE-defined DHSs for individual Patski and GM12878 cells.

Fig. 2

Fig. 2. Single-cell ATAC-seq deconvolutes human cell type mixtures

(A–C): GM12878/HEK293T nuclei. (D–F): GM12878/HL-60 nuclei. (A & D) Histograms of proportions of reads mapping to cell-type specific DHSs that correspond to one cell type or the other. (B & E) Boxplots of the overall fraction of reads mapping to ENCODE-defined DHSs for individual cells. (C & F) Multidimensional scaling of single-cell ATAC-seq data using pairwise Jaccard distances between cells based on DHS usage. Cell type assignments based on proportions shown in A & D.

Fig. 3

Fig. 3. Single-cell ATAC-seq identifies functionally relevant differences in accessibility between cell types

(A) Bar plot for relative fraction of DHSs overlapping each chromatin state (HL-60 vs. GM12878). Gray bars show frequencies for all sites tested. Blue bars show frequencies for differentially accessible sites. CTCF=CTCF enriched element; E=Predicted enhancer; PF=Predicted promoter flanking region; R=Predicted repressed; T=Predicted transcribed; TSS=Predicted promoter region; WE=Predicted weak enhancer. *=significant difference in proportions. Values do not add to 1 because sites can overlap multiple chromatin states. (B) Multidimensional scaling of chromatin accessibility data for 14,533 cells (GM12878/HL-60 mixtures from 13 experiments on 4 dates). (C) Heatmap of hypersensitive site usage for 10,241 cells (columns) at 21,378 DHSs (rows) (GM12878/HL-60 mixtures). Colors indicate accessibility of sites after latent semantic indexing. Top color bar is coded by cell-type assignments (green=HL-60; blue=GM12878; black=unassigned). Left color bar indicates modules formed by clustering DHSs.

Fig. 4

Fig. 4. Single-cell ATAC-seq identifies GM12878 subtypes

(A) Heatmap of chromatin accessibility measures after latent semantic indexing of DHS usage shows GM12878 cells cluster into subpopulations. Modules of coordinately accessible chromatin accessibility are significantly enriched for binding of selected transcription factors (TFs) (examples on right). (B) Detailed depiction of LYN locus. Top shows “co-accessibility scores” between the transcription start sites and four putative enhancers in the region, which are Pearson correlation values of LSI accessibility scores between cells, for six DHSs present in this region. Height and thickness of each loop indicates the strength of correlation (red=positive; blue=negative). Middle shows in which subtypes (defined in top bar of (A)) these elements are most often accessible. Bottom shows ENCODE data for this region from UCSC browser, including transcript model, DHS peaks, ChIP-seq binding profiles for several TFs, and predicted chromatin state.

Comment in

Similar articles

Cited by

References

    1. Stergachis AB, et al. Cell. 2013;154:888–903. - PMC - PubMed
    1. Thurman RE, et al. Nature. 2012;489:75–82. - PMC - PubMed
    1. Boyle AP, et al. Cell. 2008;132:311–22. - PMC - PubMed
    1. Buenrostro JD, Giresi PG, Zaba LC, Chang HY, Greenleaf WJ. Nat Methods. 2013;10:1213–8. - PMC - PubMed
    1. Navin N, et al. Nature. 2011;472:90–4. - PMC - PubMed

Publication types

MeSH terms

Substances

LinkOut - more resources