CoChat
Sign Up
Cited by 109Genomics and Chromatin DynamicsChromatin Remodeling and CancerCRISPR and Genetic Engineering

Profiling the genetic determinants of chromatin accessibility with scalable single-cell CRISPR screens

Open access
Liscovitch-Brauer, Noa, Montalbano, Antonino, Deng, Jiale, Méndez-Mancilla, Alejandro, Wessels, Hans-Hermann, Moss, Nicholas G., Kung, Chia-Yu, Sookdeo, Akash +5 more·2021·Nature Biotechnology
DOI 10.1038/s41587-021-00902-xPMC PMC8516442

Published version of a preprint

Open this paper in CoChat — free accountOpen PDF

Abstract

CRISPR screens have been used to connect genetic perturbations with changes in gene expression and phenotypes. Here we describe a CRISPR-based, single-cell combinatorial indexing assay for transposase-accessible chromatin (CRISPR-sciATAC) to link genetic perturbations to genome-wide chromatin accessibility in a large number of cells. In human myelogenous leukemia cells, we apply CRISPR-sciATAC to target 105 chromatin-related genes, generating chromatin accessibility data for ~30,000 single cells. We correlate the loss of specific chromatin remodelers with changes in accessibility globally and at the binding sites of individual transcription factors (TFs). For example, we show that loss of the H3K27 methyltransferase EZH2 increases accessibility at heterochromatic regions involved in embryonic development and triggers expression of genes in the HOXA and HOXD clusters. At a subset of regulatory sites, we also analyze changes in nucleosome spacing following the loss of chromatin remodelers. CRISPR-sciATAC is a high-throughput, single-cell method for studying the effect of genetic perturbations on chromatin in normal and disease states.

Integrity signals

Registry

Crossref · 10.1038/s41587-021-00902-x

Cites

55 of 55 references verified0 citation errors · 0 not checkable

Statistics

Nothing to recompute.

Claims the paper makes

0 of 17 cited sources checked · 0 supported · 0 flagged

  • Findingown data

    Loss of EZH2 triggers expression of genes in the HOXA and HOXD clusters.

    “and triggers expression of genes in the HOXA and HOXD clusters.”
    Quote in the paperSubstantial claim
  • Findingown data

    Loss of specific chromatin remodelers correlates with changes in accessibility globally and at binding sites of individual transcription factors.

    “We correlate loss of specific chromatin remodelers with changes in accessibility globally and at the binding sites of individual transcription factors.”
    Quote in the paperSubstantial claim
  • Findingown data

    In human myelogenous leukemia cells, CRISPR-sciATAC targeting 105 chromatin-related genes produced chromatin accessibility data for about 30,000 single cells.

    “In human myelogenous leukemia cells, we apply CRISPR-sciATAC to target 105 chromatin-related genes, generating chromatin accessibility data for ~30,000 single cells.”
    Quote in the paperSubstantial claim
  • Findingown data

    Loss of EZH2 increases accessibility at heterochromatic regions involved in embryonic development.

    “we show that loss of the H3K27 methyltransferase EZH2 increases accessibility at heterochromatic regions involved in embryonic development”
    Quote in the paperSubstantial claim
  • Findingown data

    In a mix of human HEK293 and mouse NIH3T3 cells, 93% of cell barcodes had gRNA-containing reads uniquely assigned to either human or mouse gRNAs.

    “We found that 93% of cell barcodes had gRNA-containing reads that could uniquely be assigned to either human or mouse gRNAs”
    Quote in the paperSubstantial claim
  • The quote only partly states this claim. 1
  • May combine two claims. 12
  • Classified as probably not a substantial claim. 8

  • The quote only partly states this claim.

    Findingown data

    At a subset of regulatory sites, loss of chromatin remodelers changes nucleosome spacing.

    “At a subset of regulatory sites, we also analyze changes in nucleosome spacing upon loss of chromatin remodelers.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    CRISPR-sciATAC can generate thousands of single cells at about 20x less reagent cost and requires about 14x less time compared with Perturb-ATAC.

    “When compared with Perturb-ATAC, CRISPR-sciATAC can generate thousands of single cells at ~20x less reagent cost and requires ~14x less time (Supplementary Tables 8 and 9).”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    TnY has insertion site preferences distinct from, but of a similar magnitude to those of Tn5.

    “we found that TnY has insertion site preferences distinct from, but of a similar magnitude to those of Tn5”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    Loss of SWI/SNF-ATPase subunit ARID1A or ISWI-ATPase subunit SMARCA5 results in many changes in TFBS accessibility.

    “Loss of SWI/SNF-ATPase subunit ARID1A or ISWI-ATPase subunit SMARCA5 results in many changes in TFBS accessibility”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    Knock-out of CoREST subunit SFMBT1 tends to cause nucleosome expansion around TFBSs in promoters but not in enhancers, for the 7 TFs analyzed; for example, an 85 nt expansion around AP-1 binding sites in promoters and no change in nucleosomal positions around AP-1 binding sites in enhancers.

    “Knock-out of CoREST subunit SFMBT1 tends to cause nucleosome expansion around TFBSs in promoters but not in enhancers, for the 7 TFs analyzed: for example, an 85 nt expansion around AP-1 binding sites in promoters and no change in nucleosomal positions around AP-1 binding sites in enhancers”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    In this work, 28,510 cells were analyzed, which is about 7-fold more cells than in the Perturb-ATAC dataset.

    “In this work, we analyzed 28,510 cells, which is ~7-fold more cells than in the Perturb-ATAC dataset”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    After filtering for cells with at least 500 unique ATAC-seq fragments and at least 100 gRNA reads, 11,104 cells were obtained with a median of 1,977 unique ATAC-seq fragments mapping to the human genome.

    “After filtering for cells with ≥500 unique ATAC-seq fragments and ≥100 gRNA reads (Supplementary Fig. 3c–h), we obtained 11,104 cells with a median of 1,977 unique ATAC-seq fragments mapping to the human genome.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    Among the 3 gRNAs targeting EZH2, the least effective gRNA resulted in less HOX gene expression, reinforcing the role of EZH2 in maintaining HOX gene repression.

    “Among the 3 gRNAs targeting EZH2, we noticed that the least effective gRNA resulted in less HOX gene expression, further reinforcing the role of EZH2 in maintaining HOX gene repression”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    All 66 gRNAs were recovered with a median of 148 single cells per gRNA and 468 single cells per gene.

    “We recovered all 66 gRNAs with a median of 148 single cells per gRNA and 468 single cells per gene.”
    Quote in the paperSubstantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation

    Alterations in chromatin state are associated with many diseases, including several cancers.

    “Alterations in chromatin state have been associated with many diseases including several cancers1.”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [1] Flavahan, W. A., Gaskell, E. & Bernstein, B. E. Epigenetic plasticity and the hallmarks of cancer. Science 357, eaal2380 (2017). source

      Not checked in the free run.

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation

    CRISPR screens combined with single-cell RNA-sequencing measure effects of genetic perturbations on gene expression across the transcriptome.

    “In recent years, CRISPR screens have been combined with single-cell RNA-sequencing to measure the effects of genetic perturbations on gene expression across the transcriptome2–5.”
    Quote in the paperNot a substantial claim0 of 4 sources checked
    • cites [2] Datlinger, P. et al. Pooled CRISPR screening with single-cell transcriptome readout. Nat. Methods 14, 297–301 (2017). source

      Not checked in the free run.

    • cites [3] Adamson, B. et al. A multiplexed single-cell CRISPR screening platform enables systematic dissection of the unfolded protein response. Cell 167, 1867–1882 (2016). source

      Not checked in the free run.

    • cites [4] Dixit, A. et al. Perturb-seq: dissecting molecular circuits with scalable single-cell RNA profiling of pooled genetic screens. Cell 167, 1853–1866 (2016). source

      Not checked in the free run.

    • cites [5] Jaitin, D. A. et al. Dissecting immune circuits by linking CRISPR-pooled screens with single-cell RNA-seq. Cell 167, 1883–1896 (2016). source

      Not checked in the free run.

  • May combine two claims.

    Study design and conductnone given

    One important caveat of CRISPR nuclease-driven perturbation is that knock-out can be incomplete due to in-frame repair and efficiency can vary depending on the gRNA.

    “one important caveat of CRISPR nuclease-driven perturbation is that knock-out can be incomplete due to in-frame repair and efficiency can vary depending on the gRNA.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S2 (checkable). May combine two claims.

    Interpretationreasoning

    Overall, CRISPR-sciATAC can be applied to study diverse phenotypes and diseases and to understand the interaction between genetic changes and genome-wide chromatin accessibility.

    “Overall, CRISPR-sciATAC can be applied to study diverse phenotypes and diseases and to understand the interaction between genetic changes and genome-wide chromatin accessibility.”
    Quote in the paperNot a substantial claim

  • Classified as probably not a substantial claim.

    Study design and conductcitation

    P-values were adjusted for multiple hypothesis testing using a Benjamini-Hochberg false-discovery rate correction with a threshold of q ≤ 0.1.

    “The p-values were adjusted for multiple hypothesis testing using a Benjamini-Hochberg false-discovery rate correction (q ≤ 0.1).”
    • Also stated in: § Statistical analysis · plain
      “Corrections for multiple-hypothesis testing was performed using the Benjamini-Hochberg approach48.”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites [48] Benjamini, Y. & Hochberg, Y. Controlling the false discovery rate: a practical and powerful approach to multiple testing. J. R. Stat. Soc. Ser. B Stat. Methodol. 57, 289–300 (1995). source

      Not checked in the free run.

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation

    The transcription factors analyzed have previously been shown to have a bimodal nucleosome profile.

    “These TFs have also been previously shown to have a bimodal profile24.”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [24] Kundaje, A. et al. Ubiquitous heterogeneity and asymmetry of the chromatin environment at regulatory elements. Genome Res. 22, 1735–1747 (2012). source

      Not checked in the free run.

  • May combine two claims.

    Interpretationreasoning

    CRISPR-sciATAC does not depend critically on any specific gRNA but rather looks for consistent effects between gRNAs.

    “However, CRISPR-sciATAC does not depend critically on any specific gRNA but rather looks for consistent effects between gRNAs.”
    Quote in the paperSubstantial claim

  • Classified as probably not a substantial claim.

    Study design and conductreasoning

    Empirical p-values for nucleosome expansion or compaction scores were generated with a label-permutation test that shuffled non-targeting and knock-out labels while keeping group sizes constant, to avoid bias from unequal cell numbers.

    “Empirical p-values for each score were generated using a label-permutation test, where non-targeting and knock-out labels were randomly shuffled while keeping group size constant to avoid biases that stem from different numbers of cells in the non-targeting and knock-out cell populations.”
    Quote in the paperSubstantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductnone given

    The gRNA library coverage was 1,500x on average for the species-mixing experiment, the chromatin modifier screen and the chromatin remodeling complex subunit screen.

    “The gRNA library coverage was 1,500x on average for the species-mixing experiment, the chromatin modifier screen and the chromatin remodeling complex subunit screen.”
    Quote in the paperNot a substantial claim

  • Classified as probably not a substantial claim. May combine two claims.

    Study design and conductreasoning

    To avoid bias from differing numbers of recovered cells per gRNA, all gRNA cell populations were subsampled to 12 cells and the deviation Z-score calculation was repeated 1,000 times.

    “To avoid biases from recovery of different numbers of cells for each gRNA, we subsampled all gRNA cell populations to 12 cells (the lowest number of cells for a single gRNA in our K562 dataset), calculated the deviation Z-scores, and repeated this resampling process 1000 times to obtain deviation Z-scores for each gRNA.”
    Quote in the paperSubstantial claim
19 of 19

  • The paper relies on this without support.

    Chromatin accessibility controls gene expression and is dynamically regulated during differentiation and homeostasis.

    “Chromatin accessibility orchestrates cis- and trans-regulatory interactions to control gene expression and is dynamically regulated in cell differentiation and homeostasis.”

  • The paper relies on this without support.

    CRISPR nuclease-induced loss of a chromatin modifier in K562 cells models loss-of-function mutations in cancer.

    “Taken together, these results suggest that loss-of-function mutations in EZH2 lead to aberrant expression of genes from HOXA and HOXD clusters.”

  • The paper relies on this without support.

    The presence of a gRNA in a cell indicates functional perturbation of the targeted gene.

    “CRISPR-sciATAC does not depend critically on any specific gRNA but rather looks for consistent effects between gRNAs.”

  • The paper relies on this without support.

    ENCODE ChIP-seq data from K562 accurately represents the chromatin marks in the studied K562 cells.

    “We examined how loss of these chromatin modifiers impacts accessibility within known chromatin marks (primarily histone post-translation modifications) using ENCODE ChIP-seq data from K562 cells”

  • The paper relies on this without support.

    Non-targeting gRNAs have no effect on chromatin accessibility.

    “For cells receiving a non-targeting gRNA, we find that 75 cells correlate well with the respective pseudo-bulk populations”
5 of 5

Claims other papers make about this one

Each is checked against this paper’s own text: does it say what the citing paper says it does?

  • Five methods commonly in use for low-MOI single-cell CRISPR screen association testing are Seurat-Wilcox, MIMOSCA, t-test, MAST, and KS test.

    “identified five methods commonly in use: the default Seurat [19] FindMarkers() function based on the Wilcoxon test (Seurat-Wilcox), MIMOSCA [1], a t-test on the library-size-normalized expressions [10], MAST [20], and a Kolmogorov-Smirnov (KS) test on the library-size-normalized expressions [21]”

    From Robust differential expression testing for single-cell CRISPR screens at low multiplicity of infection (2024) · cites this paper as [10]

    The citation is attached to the t-test method, not to the entire list of five methods. The Methods sections on Statistical analysis and Differential accessibility at genomic regions with specific chromatin and DNA modifications report using t-tests, but on group data and chromatin-accessibility Z-scores, respectively—not on library-size-normalized gene expression. No page number is available in our copy. (LLM, conf=0.85)

    Experimental — accuracy not yet measured

  • The authors included scATAC-seq data from three different sources: Spear-ATAC, CRISPR-sciATAC, and ASAP-seq.

    “We included scATAC-seq data from three different sources: Spear-ATAC42, CRISPR-sciATAC59 and ASAP-seq39.”

    From scPerturb: harmonized single-cell perturbation data (2024) · cites this paper as [59]

  • CRISPR-sciATAC enables joint capture of chromatin accessibility profiles and CRISPR perturbations.

    “A similar approach involves CRISPR screens with CRISPR–sciATAC [21], which enables the joint capture of chromatin accessibility profiles and CRISPR perturbations.”

    From A mini-review on perturbation modelling across single-cell omic modalities (2024) · cites this paper as [21]

  • Compared to Perturb-ATAC, both Spear-ATAC and CRISPR-sciATAC offer higher throughput.

    “Compared to Perturb-ATAC, both Spear-ATAC and CRISPR–sciATAC offer higher throughput.”

    From A mini-review on perturbation modelling across single-cell omic modalities (2024) · cites this paper as [21]

  • CRISPR-sciATAC can generate thousands of single cells at approximately 20-fold less reagent cost and requires approximately 14-fold less time than Perturb-ATAC.

    “Furthermore, CRISPR–sciATAC can generate thousands of single cells at ∼20 × less reagent cost and requires ∼14 × less time than Perturb-ATAC.”

    From A mini-review on perturbation modelling across single-cell omic modalities (2024) · cites this paper as [21]

Cites

Flagged first, then by citations

Cited by

Sorted by citations

Related

Authors
Liscovitch-Brauer, N., Montalbano, A., Deng, J., Méndez-Mancilla, A., Wessels, H.-H., Moss, N. G., Kung, C.-Y., Sookdeo, A., Guo, X., Geller, E., Jaini, S., Smibert, P., & Sanjana, N. E.
Affiliations
New York Genome Center (US)
Corresponding author
Sanjana, Neville E.
ORCID
Venue
Nature Biotechnology
Publisher
Springer Science and Business Media LLC
Published
2021-04-29
Volume
39
Issue
10
Pages
1270-1277
ISSN
  • 0733-222X (ISSN-L)
  • 1087-0156
  • 1546-1696
DOI
10.1038/s41587-021-00902-x
PMC
PMC8516442
OpenAlex
W3158694246
Cited by
107
References
52
Citations by year
2026: 10 · 2025: 31 · 2024: 30 · 2023: 14 · 2022: 17 · 2021: 5
FWCI
5.49
Type
journal article
Language
en
Open access
Licence
Locations
  • Nature Biotechnology (publishedVersion)
  • bioRxiv (Cold Spring Harbor Laboratory) (acceptedVersion, other-oa)
  • PubMed (publishedVersion)
  • PubMed Central (acceptedVersion)
Funders
  • Human Frontier Science Program (LT000672/2019-L)
  • European Molecular Biology Organization (ALTF 826-2018)
  • U.S. Department of Health & Human Services | National Institutes of Health (DP2HG010099, R01CA218668)
Topics
  • Life Sciences › Biochemistry, Genetics and Molecular Biology › Molecular Biology › Genomics and Chromatin Dynamics
  • Life Sciences › Biochemistry, Genetics and Molecular Biology › Molecular Biology › Chromatin Remodeling and Cancer
  • Life Sciences › Biochemistry, Genetics and Molecular Biology › Molecular Biology › CRISPR and Genetic Engineering
Keywords
chromatin accessibility, single-cell CRISPR screens, chromatin remodelers, EZH2, H3K27 methylation, heterochromatin, transcription factor binding sites, nucleosome spacing, myeloid leukemia
MeSH terms
RNA-Seq, Binding Sites, Chromatin, Humans, Leukemia, Myeloid, Nucleosomes, Transcription Factors, Transposases, Gene Expression Profiling, Chromatin Assembly and Disassembly, Regulatory Elements, Transcriptional, Epigenomics, Single-Cell Analysis, Clustered Regularly Interspaced Short Palindromic Repeats

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Claims read by deepseek-v4.1-flash under definition version 3 (extract-v3) on 2026-10-04.

Classified by jev-1.13-20260917 (gold-guideline-v0.2) on 2026-10-04.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review