CoChat
Sign Up
Cited by 483Single-cell and spatial transcriptomicsGenomics and Chromatin DynamicsEpigenetics and DNA Methylation

Spatial epigenome–transcriptome co-profiling of mammalian tissues

Open access
Zhang, Di, Deng, Yanxiang, Kukanja, Petra, Agirre, Eneritz, Bartosovic, Marek, Dong, Mingze, Ma, Cong, Ma, Sai +15 more·2023·Nature
DOI 10.1038/s41586-023-05795-1PMC PMC10076218

Published version of a preprint

Open this paper in CoChat — free accountOpen PDF

Abstract

Emerging spatial technologies, including spatial transcriptomics and spatial epigenomics, are becoming powerful tools for profiling of cellular states in the tissue context 1–5 . However, current methods capture only one layer of omics information at a time, precluding the possibility of examining the mechanistic relationship across the central dogma of molecular biology. Here, we present two technologies for spatially resolved, genome-wide, joint profiling of the epigenome and transcriptome by cosequencing chromatin accessibility and gene expression, or histone modifications (H3K27me3, H3K27ac or H3K4me3) and gene expression on the same tissue section at near-single-cell resolution. These were applied to embryonic and juvenile mouse brain, as well as adult human brain, to map how epigenetic mechanisms control transcriptional phenotype and cell dynamics in tissue. Although highly concordant tissue features were identified by either spatial epigenome or spatial transcriptome we also observed distinct patterns, suggesting their differential roles in defining cell states. Linking epigenome to transcriptome pixel by pixel allows the uncovering of new insights in spatial epigenetic priming, differentiation and gene regulation within the tissue architecture. These technologies are of great interest in life science and biomedical research.

AI snapshot

Written from the paper’s full text.

TL;DR

The paper presents spatial ATAC–RNA-seq and spatial CUT&Tag–RNA-seq for genome-wide joint profiling of chromatin accessibility or histone modifications and mRNA on the same tissue section at near-single-cell resolution. These methods were applied to embryonic and juvenile mouse brain and adult human hippocampus, revealing spatial epigenetic–transcriptional relationships including epigenetic priming and region-specific gene regulation.

Key findings

  • In P22 mouse brain with 100 × 100 barcodes, spatial ATAC–RNA-seq yielded a median of 14,284 unique fragments per pixel, with 19% enriched at transcription start sites and 26% in peaks; RNA detected 22,914 genes, averaging 1,073 genes and 2,358 UMIs per pixel.
  • Spatial CUT&Tag–RNA-seq on P22 mouse brain gave median unique fragments per pixel of 10,644 for H3K27me3, 10,002 for H3K27ac and 2,507 for H3K4me3; paired RNA averaged 2,011, 1,513 and 1,329 genes per pixel, respectively.
  • Joint clustering of E13 mouse embryo ATAC and RNA identified a new neuronal cluster J10 not readily resolved by single modalities; in P22 mouse brain, 21,417 significant peak-to-gene linkages were detected by ATAC–RNA-seq and 19,468 from H3K27ac CUT&Tag–RNA-seq.
  • Region-specific regulation was observed: H3K27me3 anticorrelated with RNA in corpus callosum (for example, Mal, Mag and Car2 low CSS/high RNA; Grin2b and Syt1 high CSS/low RNA), while H3K27ac and H3K4me3 correlated with RNA; potential bivalency/poising was noted for genes such as Ptprz1 and Fnbp1. In human hippocampus, PROX1 was highly expressed in the granule cell layer but showed modest chromatin accessibility.

Methods

  • Frozen tissue sections were fixed, subjected to Tn5 transposition for ATAC or antibody-guided pA-Tn5 CUT&Tag, then in-tissue reverse transcription; sequential microfluidic barcode A and B ligation generated 50 × 50 or 100 × 100 pixel grids (2,500 or 10,000 pixels), with separate gDNA and cDNA libraries sequenced on Illumina NovaSeq 6000 (paired-end, 150 bp).
  • Samples included E13 mouse embryo, P21/22 mouse brain and adult human hippocampus from a 31-year-old male; pixel sizes were 50 μm for embryo and human tissue and 20 μm for juvenile mouse brain. Analyses used Signac, Seurat and ArchR, with label transfer to scRNA-seq, scATAC-seq and scCUT&Tag atlases, gene activity score (GAS), chromatin silencing score (CSS), peak-to-gene links, pseudotime and GO enrichment.
  • Reproducibility was high: Pearson correlations were r = 0.98 for ATAC and r = 0.98 for RNA in spatial ATAC–RNA-seq, and r = 0.96 for CUT&Tag (H3K27ac) and r = 0.89 for RNA in spatial CUT&Tag–RNA-seq.

Limitations

  • The paper states that low RNA expression despite open chromatin may be due in part to sequencing depth and RNA detection sensitivity, and interprets some such cases as lineage priming.
  • The method is near-single-cell, not single-cell: 50-μm pixels contained an average of 25 cells for E10 mouse embryo and 1–9 cells for human hippocampus, while 20-μm pixels contained 1–3 cells in juvenile mouse brain; when more than one cell type was present in a pixel, the major cell type was assigned.
  • Evident from the text: the human analysis used one 31-year-old male hippocampus sample, limiting generalizability, and the epigenome–transcriptome relationships are largely correlative rather than causally tested.

Model-written summary. Check it against the paper before relying on it.

Integrity signals

Registry

Crossref · 10.1038/s41586-023-05795-1

Cites

65 of 65 references verified0 citation errors · 0 not checkable

Statistics

Nothing to recompute.

Claims the paper makes

0 of 47 cited sources checked · 0 supported · 0 flagged

  • Findingown data

    Most 20-μm pixels contained one to three cells per pixel in juvenile mouse brain.

    “Most 20-μm pixels contained one to three cells per pixel in juvenile mouse brain (Supplementary Fig. 1b).”
    Quote in the paperNot a substantial claim
  • Findingown data

    21,417 significant peak-to-gene linkages between regulatory elements and target genes were detected.

    “We detected 21,417 significant peak-to-gene linkages between regulatory elements and target genes”
    Quote in the paperSubstantial claim
  • Findingown data

    The loss of chromatin accessibility at the Pax6 and radial glia marker Fabp7 loci preceded downregulation of the corresponding RNAs.

    “The loss of chromatin accessibility at the Pax6 and radial glia marker Fabp7 loci preceded downregulation of the corresponding RNAs”
    Quote in the paperSubstantial claim
  • Findingown data

    CUT&Tag and RNA showed good concordance in spatial patterns.

    “These clusters agreed with the anatomical annotation by Nissl staining and showed good concordance between CUT&Tag and RNA in spatial patterns (Fig. 3a–c), which was further confirmed by Belayer34 analysis (Supplementary Fig. 7c).”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites [34] Ma, C., Chitra, U., Zhang, S. & Raphael, B. J. Belayer: modeling discrete and continuous spatial variation in gene expression from spatially resolved transcriptomics. Cell Syst. 13, 786–797 (2022). source

      Not checked in the free run.

  • Findingown data

    The CUT&Tag and RNA clusters agreed with anatomical annotation by Nissl staining.

    “These clusters agreed with the anatomical annotation by Nissl staining and showed good concordance between CUT&Tag and RNA in spatial patterns (Fig. 3a–c), which was further confirmed by Belayer34 analysis (Supplementary Fig. 7c).”
    Quote in the paperSubstantial claim
  • Classified as probably not a substantial claim. 13
  • May combine two claims. 46

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

    Findingnone given

    Using a 100 × 100 barcode scheme, the mapping area covers nearly the entire hemisphere of a P22 mouse brain coronal section.

    “Using a 100 × 100 barcode scheme, the mapping area covers nearly the entire hemisphere of a P22 mouse brain coronal section.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    Spatial CUT&Tag-RNA-seq for H3K27me3, H3K27ac and H3K4me3 on mouse P21/22 brains yielded median unique fragments per pixel of 10,644, 10,002 and 2,507, respectively, with TSS overlap of 12%, 17% and 67% and peak overlap of 12%, 21% and 54%.

    “we obtained a median of 10,644 (H3K27me3), 10,002 (H3K27ac) and 2,507 (H3K4me3) unique fragments per pixel, of which 12% (H3K27me3), 17% (H3K27ac) and 67% (H3K4me3) overlapped with transcription start site regions and 12% (H3K27me3), 21% (H3K27ac) and 54% (H3K4me3) were located in peaks”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    For the RNA data of spatial CUT&Tag-RNA-seq, 25,881 (H3K27me3), 23,415 (H3K27ac) and 22,731 (H3K4me3) genes were detected with an average of 2,011, 1,513 and 1,329 genes per pixel and 4,734, 3,580 and 2,885 UMIs per pixel, respectively.

    “For the RNA data, 25,881 (H3K27me3), 23,415 (H3K27ac) and 22,731 (H3K4me3) genes were detected with an average of 2,011 (H3K27me3), 1,513 (H3K27ac) and 1,329 (H3K4me3) genes per pixel (4,734 (H3K27me3), 3,580 (H3K27ac) and 2,885 (H3K4me3) UMIs per pixel, respectively”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    For the P22 mouse brain sample, a median of 14,284 unique fragments per pixel was obtained, of which 19% were enriched in transcription start site regions and 26% located in peaks.

    “From this sample we obtained a median of 14,284 unique fragments per pixel, of which 19% were enriched in the transcription start site regions and 26% located in peaks”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    For 50-μm pixel size, an average of 25 cells was obtained for E10 mouse embryo and between one and nine cells for human hippocampus.

    “For the 50-μm pixel size we obtained an average of 25 cells for E10 mouse embryo1 and between one and nine cells for human hippocampus3.”
    Quote in the paperNot a substantial claim0 of 2 sources checked
    • cites [1] Liu, Y. et al. High-spatial-resolution multi-omics sequencing via deterministic barcoding in yissue. Cell 183, 1665–1681 (2020). source

      Not checked in the free run.

    • cites [3] Deng, Y. et al. Spatial profiling of chromatin accessibility in mouse and human tissues. Nature 609, 375–383 (2022). source

      Not checked in the free run.

  • May combine two claims.

    Findingown data

    Spatial epigenome and spatial transcriptome identified highly concordant tissue features but also distinct patterns.

    “Although highly concordant tissue features were identified by either spatial epigenome or spatial transcriptome we also observed distinct patterns”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    For the RNA portion of spatial ATAC-RNA-seq, a total of 22,914 genes were detected, with an average of 1,073 genes and 2,358 UMIs per pixel.

    “For the RNA portion a total of 22,914 genes were detected, with an average of 1,073 genes and 2,358 unique molecular identifiers (UMIs) per pixel (spatial ATAC–RNA-seq)”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingcitation

    Immature oligodendrocytes and oligodendrocytes were observed in cluster A3 corresponding to corpus callosum; a thin layer of radial glia-like cells was found in the lateral ventricle, medium spiny neurons (D1MSN and D2MSN) in striatum and inhibitory neurons in the lateral septal nucleus based on chromatin accessibility.

    “we observed immature oligodendrocytes and oligodendrocytes in cluster A3 corresponding to corpus callosum. A thin layer of radial glia-like cells was found in the lateral ventricle, medium spiny neurons (D1MSN and D2MSN) in striatum and inhibitory neurons in the lateral septal nucleus30”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites [30] Li, Y. E. et al. An atlas of gene regulatory elements in adult mouse cerebrum. Nature 598, 129–136 (2021). source

      Not checked in the free run.

  • May combine two claims.

    Findingown data

    The study identified 13 and 15 specific clusters for H3K27me3 and RNA, 12 and 13 clusters for H3K27ac and RNA, and 11 and 12 clusters for H3K4me3 and RNA, respectively.

    “We identified 13 and 15 specific clusters for H3K27me3 and RNA, 12 and 13 clusters for H3K27ac and RNA and 11 and 12 clusters for H3K4me3 and RNA, respectively (Fig. 3a–c).”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    Monocle2 pseudotime analysis showed a bifurcation in chromatin accessibility but not in RNA expression; one path led to regions close to the ventricular zone and the other terminated in regions distal to the ventricular zone.

    “Monocle2 pseudotime analysis showed a bifurcation in chromatin accessibility but not in RNA expression; one path led to regions close to the ventricular zone (green pixels, Extended Data Fig. 3d,e) and the other terminated in regions distal to the ventricular zone (blue pixels).”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    Chromatin accessibility of marker genes identified major regions such as striatum (Pde10a and Adcy5, cluster A1), corpus callosum (Sox10, Mbp and Tspan2, cluster A3), cortex (Mef2c, Neurod6 and Nrn1, clusters A0 and A4; Cux2, cluster A4) and lateral ventricle (Dlx1, Pax6, Notch1 and Sox2, cluster A11).

    “Chromatin accessibility of marker genes identified major regions such as striatum (Pde10a and Adcy5, markers of medium spiny neurons, cluster A1), corpus callosum (Sox10, Mbp and Tspan2, markers of oligodendroglia, cluster A3), cortex (Mef2c, Neurod6 and Nrn1, clusters A0 and A4, markers of excitatory neurons; Cux2, cluster A4) and lateral ventricle (Dlx1, Pax6, Notch1 and Sox2, markers of ependym”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    Spatial ATAC-RNA-seq on E13 mouse embryo identified eight major ATAC clusters and 14 RNA clusters.

    “Spatial ATAC–RNA-seq on E13 mouse embryo identified eight major ATAC clusters and 14 RNA clusters”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    Genes involved in neuronal identity, axonogenesis and synapse organization, such as Dcx and Tubb3, showed increased expression in spatial pseudotime but chromatin accessibility at their loci was already elevated at earlier stages.

    “genes involved in neuronal identity, axonogenesis and synapse organization (GO Myt1l to Dnm3 in Extended Data Fig. 3a), such as Dcx and Tubb3, showed increased expression in spatial pseudotime but chromatin accessibility at their loci was already elevated at earlier stages”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    Radial glia (neural stem/progenitor cells) were observed predominantly in the ventricular layer, and differentiated cell types such as postmitotic premature neurons and inhibitory interneurons were enriched in regions distant from the ventricular layer.

    “radial glia (neural stem/progenitor cells) were observed predominantly in the ventricular layer, and differentiated cell types such as postmitotic premature neurons and inhibitory interneurons were enriched in those regions distant from the ventricular layer”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    Spatial RNA clusters showed region-specific gene expression such as striatum (Pde10a, cluster R2, medium spiny neurons), corpus callosum (Mbp and Tspan2, cluster R5, oligodendroglia) and cortex (Mef2c, clusters R0 and R1, excitatory neurons).

    “Spatial RNA clusters also showed region-specific gene expression such as striatum (Pde10a, cluster R2, medium spiny neurons), corpus callosum (Mbp and Tspan2, cluster R5, oligodendroglia) and cortex (Mef2c, clusters R0 and R1, excitatory neurons)”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    Satb2 showed the highest activity in clusters C8 and C9 (H3K27ac) and C6 and C7 (H3K4me3), but the lowest CSS in clusters C1 and C3 (H3K27me3) in the cortical layer.

    “Satb2 showed the highest activity in clusters C8 and C9 (H3K27ac) and C6 and C7 (H3K4me3), but the lowest CSS in clusters C1 and C3 (H3K27me3) in the cortical layer.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    In the superficial cortical layer, genes for GABAergic regulation of synaptic transmission were enriched in H3K27me3 whereas glutamatergic synapse transmission genes had high expression and low H3K27me3.

    “For instance, in the superficial cortical layer, genes for GABAergic regulation of synaptic transmission were enriched in H3K27me3 whereas glutamatergic synapse transmission genes had high expression and low H3K27me3 (Supplementary Figs. 9b, 10b and 11b; GOs for striatum, superficial and deeper cortical layers).”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    TF motifs (NEUROD1 and SNAI1) and their spatial patterns were visualized in different tissue regions, and positive TF regulators were identified.

    “TF motifs (NEUROD1 and SNAI1) and their spatial patterns were visualized in different tissue regions, and positive TF regulators were identified (Extended Data Fig. 10b,c).”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    These techniques were applied to comapping of embryonic and juvenile mouse brain, as well as adult human brain.

    “These techniques were applied to comapping of embryonic and juvenile mouse brain, as well as adult human brain.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    The study identified seven and eight major clusters for ATAC and RNA, respectively, and the spatial pattern agreed with major anatomical landmarks in adult human brain hippocampal formation.

    “We identified seven and eight major clusters for ATAC and RNA, respectively, and the spatial pattern agreed with major anatomical landmarks (Fig. 5a)3, including specific marker genes (Fig. 5d).”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    PROX1 is highly expressed in GCL but showed modest chromatin accessibility.

    “For example, PROX1, a signature gene defining granule neuron identity during pyramidal neuron fate selection44, is indeed highly expressed in GCL but showed modest chromatin accessibility (Fig. 5d).”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    ATAC cluster A4 represents the granule cell layer (THY1, BCL11B) and cluster A6 corresponds to the choroid plexus.

    “The ATAC cluster A4 represents the granule cell layer (GCL) (THY1, BCL11B) and cluster A6 corresponds to the choroid plexus (Fig. 5a,d).”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    Tspan2 was enriched in cluster C4 (H3K27ac) and cluster C1 (H3K4me3) but depleted in cluster C5 (H3K27me3) in corpus callosum.

    “Tspan2 was enriched in cluster C4 (H3K27ac) and cluster C1 (H3K4me3) but depleted in cluster C5 (H3K27me3) in corpus callosum, associated with oligodendrocyte lineage development (Supplementary Fig. 6).”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    These genes are involved in oligodendrocyte differentiation and myelination and other processes such as protein catabolism and localization.

    “These genes are involved in oligodendrocyte differentiation and myelination and other processes such as protein catabolism and localization (Supplementary Figs. 13 and 14).”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    Gpr88, a marker gene of MSN, was enriched in striatum for ATAC, H3K27ac and H3K4me3 but with low CSS in striatum for H3K27me3.

    “Gpr88, a marker gene of MSN, was enriched in striatum for ATAC, H3K27ac and H3K4me3 but with low CSS in striatum for H3K27me3 (Extended Data Fig. 9f).”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    The study observed enrichment of MOL within the corpus callosum, a thin layer of ependymal cells in the lateral ventricle, excitatory neurons in cerebral cortex and MSN in the striatum.

    “For example, we observed an enrichment of MOL within the corpus callosum, a thin layer of ependymal cells in the lateral ventricle, excitatory neurons in cerebral cortex and MSN in the striatum.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingnone given

    The study demonstrated either ATAC or CUT&Tag in conjunction with RNA for spatial multiomics mapping.

    “We have demonstrated either ATAC or CUT&Tag in conjunction with RNA for spatial multiomics mapping.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    In general H3K4me3 and H3K27ac showed high correlations, with their signals anticorrelated with H3K27me3.

    “In general H3K4me3 and H3K27ac showed high correlations, with their signals anticorrelated with H3K27me3 (Fig. 4e).”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    Granule cells were detected in the GCL, CA neurons were enriched in CA3-4 regions and VLMC were strongly distinguished in choroid plexus as opposed to other regions.

    “Granule cells (GC) were detected in the GCL, cornu ammonis (CA) neurons were enriched in CA3–4 regions and VLMC were strongly distinguished in choroid plexus as opposed to other regions.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingnone given

    Spatial epigenome-transcriptome cosequencing yielded two layers of spatial omics information directly in tissue, with data quality similar to that previously obtained by single modalities.

    “Spatial epigenome–transcriptome cosequencing yielded two layers of spatial omics information directly in tissue, with data quality similar to that previously obtained by single modalities.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    Clustering of CSS or GAS based on different modifications resolved all major tissue regions and identified region-specific marker gene modifications.

    “Clustering of CSS or GAS based on different modifications resolved all major tissue regions (Fig. 3a–c and Supplementary Fig. 6) and identified region-specific marker gene modifications.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    In contrast to H3K27me3, the study observed a robust correlation between RNA and the activating marks H3K4me3 or H3K27ac in the corpus callosum, with genes highly expressed and with high deposition in these two modalities regulating processes in oligodendroglia.

    “In contrast to H3K27me3, we observed a robust correlation between RNA and activating marks H3K4me3 or H3K27ac in the corpus callosum, with genes highly expressed and with high deposition in these two modalities regulating processes in oligodendroglia (Fig. 4b,c and Supplementary Fig. 12; GO for H3K27ac quadrant I and H3K4me3 quadrant I).”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingnone given

    The serpentine channel design allows simultaneous processing of five tissue samples, each mapped for 10,000 pixels.

    “The serpentine channel design allows simultaneous processing of five tissue samples, each mapped for 10,000 pixels.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Premisecitation

    Current spatial technologies capture only one layer of omics information at a time, precluding examination of mechanistic relationships across the central dogma.

    “However, current methods capture only one layer of omics information at a time, precluding the possibility of examining the mechanistic relationship across the central dogma of molecular biology.”
    Quote in the paperSubstantial claim0 of 5 sources checked
    • cites [1] Liu, Y. et al. High-spatial-resolution multi-omics sequencing via deterministic barcoding in yissue. Cell 183, 1665–1681 (2020). source

      Not checked in the free run.

    • cites [2] Deng, Y. et al. Spatial-CUT&Tag: spatially resolved chromatin modification profiling at the cellular level. Science 375, 681–686 (2022). source

      Not checked in the free run.

    • cites [3] Deng, Y. et al. Spatial profiling of chromatin accessibility in mouse and human tissues. Nature 609, 375–383 (2022). source

      Not checked in the free run.

    • cites [4] Chen, A. et al. Spatiotemporal transcriptomic atlas of mouse organogenesis using DNA nanoball-patterned arrays. Cell 185, 1777–1792 (2022). source

      Not checked in the free run.

    • cites [5] Lu, T., Ang, C. E. & Zhuang, X. Spatially resolved epigenomic profiling of single cells in complex tissues. Cell 185, 4448–4464 (2022). source

      Not checked in the free run.

  • May combine two claims.

    Premisecitation

    Single-cell multiomics allows uncovering mechanisms of gene regulation across different omics layers but lacks spatial information, which is crucial to understanding cellular function in tissue.

    “Single-cell multiomics allows uncovering of the mechanisms of gene regulation across different omics layers6–9 but lacks spatial information, which is crucial to understanding cellular function in tissue.”
    Quote in the paperSubstantial claim0 of 4 sources checked
    • cites [6] Allaway, K. C. et al. Genetic and epigenetic coordination of cortical interneuron development. Nature 597, 693–697 (2021). source

      Not checked in the free run.

    • cites [7] Chen, S., Lake, B. B. & Zhang, K. High-throughput sequencing of the transcriptome and chromatin accessibility in the same cell. Nat. Biotechnol. 37, 1452–1457 (2019). source

      Not checked in the free run.

    • cites [8] Cao, J. et al. Joint profiling of chromatin accessibility and gene expression in thousands of single cells. Science 361, 1380–1385 (2018). source

      Not checked in the free run.

    • cites [9] Trevino, A. E. et al. Chromatin and gene-regulatory dynamics of the developing human cerebral cortex at single-cell resolution. Cell 184, 5053–5069 (2021). source

      Not checked in the free run.

  • May combine two claims.

    Study design and conductown data

    The authors developed spatially resolved, genome-wide comapping of the epigenome and transcriptome by simultaneous profiling of chromatin accessibility and mRNA expression (spatial ATAC-RNA-seq), or histone modifications and mRNA expression (spatial CUT&Tag-RNA-seq; applied to H3K27me3, H3K27ac or H3K4me3) on the same tissue section at the cellular level via deterministic cobarcoding.

    “we developed spatially resolved, genome-wide comapping of the epigenome and transcriptome by simultaneous profiling of chromatin accessibility and messenger RNA expression (spatial assay for transposase-accessible chromatin and RNA using sequencing (spatial ATAC–RNA-seq)), or histone modifications and mRNA expression”
    Quote in the paperSubstantial claim0 of 2 sources checked
    • cites [3] Deng, Y. et al. Spatial profiling of chromatin accessibility in mouse and human tissues. Nature 609, 375–383 (2022). source

      Not checked in the free run.

    • cites [2] Deng, Y. et al. Spatial-CUT&Tag: spatially resolved chromatin modification profiling at the cellular level. Science 375, 681–686 (2022). source

      Not checked in the free run.

  • May combine two claims.

    Premisecitation

    Computational methods can integrate data from multiple omics but cannot readily uncover mechanistic links between different omics layers.

    “Although computational methods can integrate data from multiple omics10, they cannot readily uncover the mechanistic links between different omics layers.”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites [10] Ma, S. et al. Chromatin potential identified by shared single-cell profiling of RNA and chromatin. Cell 183, 1103–1116 (2020). source

      Not checked in the free run.

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

    Premisecitation

    Previously, deterministic barcoding in tissue for spatial omics sequencing was developed for spatially resolved comeasurement of the transcriptome and a panel of proteins, and recently realized in the 10X Visium platform.

    “Previously we developed deterministic barcoding in tissue for spatial omics sequencing for spatially resolved comeasurement of the transcriptome and a panel of proteins1, and recently it was also realized in the 10X Visium platform11.”
    Quote in the paperNot a substantial claim0 of 2 sources checked
    • cites [1] Liu, Y. et al. High-spatial-resolution multi-omics sequencing via deterministic barcoding in yissue. Cell 183, 1665–1681 (2020). source

      Not checked in the free run.

    • cites [11] Ben-Chetrit, N. et al. Integration of whole transcriptome spatial profiling with protein markers. Nat. Biotechnol. https://doi.org/10.1038/s41587-022-01536-3 (2023). source

      Not checked in the free run.

  • May combine two claims.

    Premisecitation

    Spatial epigenomics, transcriptomics and proteomics emerged recently but most capture only one layer of omics information.

    “Spatial epigenomics, transcriptomics and proteomics emerged recently1–5 but most of these can capture only one layer of the omics information.”
    Quote in the paperSubstantial claim0 of 5 sources checked
    • cites [1] Liu, Y. et al. High-spatial-resolution multi-omics sequencing via deterministic barcoding in yissue. Cell 183, 1665–1681 (2020). source

      Not checked in the free run.

    • cites [2] Deng, Y. et al. Spatial-CUT&Tag: spatially resolved chromatin modification profiling at the cellular level. Science 375, 681–686 (2022). source

      Not checked in the free run.

    • cites [3] Deng, Y. et al. Spatial profiling of chromatin accessibility in mouse and human tissues. Nature 609, 375–383 (2022). source

      Not checked in the free run.

    • cites [4] Chen, A. et al. Spatiotemporal transcriptomic atlas of mouse organogenesis using DNA nanoball-patterned arrays. Cell 185, 1777–1792 (2022). source

      Not checked in the free run.

    • cites [5] Lu, T., Ang, C. E. & Zhuang, X. Spatially resolved epigenomic profiling of single cells in complex tissues. Cell 185, 4448–4464 (2022). source

      Not checked in the free run.

  • May combine two claims.

    Study design and conductcitation

    Integration of single-cell ATAC-seq mouse brain atlas data with spatial ATAC-seq data facilitated identification of all major cell types, and label transfer was used to assign cell types to spatial locations where the epigenetic state may control specific cell type formation.

    “Integration of single-cell ATAC-seq mouse brain atlas data30 with spatial ATAC-seq data facilitated identification of all major cell types, and label transfer was then used to assign cell types to spatial locations where the epigenetic state may control specific cell type formation”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites [30] Li, Y. E. et al. An atlas of gene regulatory elements in adult mouse cerebrum. Nature 598, 129–136 (2021). source

      Not checked in the free run.

  • May combine two claims.

    Study design and conductnone given

    A microchip with 100 serpentine microchannels was developed to barcode 100 × 100 or, in total, 10,000 pixels per tissue section and up to five samples per chip.

    “We developed a microchip with 100 serpentine microchannels to barcode 100 × 100 or, in total, 10,000 pixels per tissue section and up to five samples per chip”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Interpretationown data

    Spatial ATAC-RNA-seq can be used to decipher the gene regulation mechanism and spatiotemporal dynamics during tissue development.

    “Thus, our spatial ATAC–RNA-seq can be used to decipher the gene regulation mechanism and spatiotemporal dynamics during tissue development.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Study design and conductown data

    Correlation analysis between replicates showed high reproducibility (r = 0.98 for ATAC, r = 0.98 for RNA in spatial ATAC-RNA-seq, r = 0.96 for CUT&Tag (H3K27ac), r = 0.89 for RNA in spatial CUT&Tag-RNA-seq).

    “Correlation analysis between replicates showed high reproducibility (r = 0.98 for ATAC, r = 0.98 for RNA in spatial ATAC–RNA-seq, r = 0.96 for CUT&Tag (H3K27ac), r = 0.89 for RNA in spatial CUT&Tag–RNA-seq”
    Quote in the paperSubstantial claim

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

    Interpretationown data

    This potential H3K4me3/H3K27me3 bivalency may reflect transcriptional poising.

    “As mentioned above, this potential H3K4me3/H3K27me3 bivalency may reflect transcriptional poising.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Interpretationown data

    Both ATAC and RNA can readily resolve all major tissue features in the human hippocampal formation, but spatial cosequencing of epigenome and transcriptome may provide new insights into dynamic gene regulation mechanisms that cannot be realized by single modalities.

    “In general, both ATAC and RNA can readily resolve all major tissue features in this region, but spatial cosequencing of epigenome and transcriptome may provide new insights into the dynamic gene regulation mechanism that cannot be realized by single modalities.”
    Quote in the paperSubstantial claim

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

    Interpretationreasoning

    The higher levels of both H3K27me3 and RNA could also be due in part to the presence in proximity of both oligodendrocyte precursor cells expressing Ptprz1 and mature oligodendrocytes where Ptprz1 is repressed in the corpus callosum.

    “This could also be due in part to the presence in proximity of both oligodendrocyte precursor cells (expressing Ptprz1) and mature oligodendrocytes (where Ptprz1 is repressed) in the corpus callosum.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Interpretationreasoning

    These technologies allow examination of spatiotemporal dynamics and genome-wide gene regulation mechanisms in the tissue context.

    “These technologies allow us to examine spatiotemporal dynamics and genome-wide gene regulation mechanisms in the tissue context.”
    Quote in the paperSubstantial claim

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

    Premisecitation

    PROX1 is a signature gene defining granule neuron identity during pyramidal neuron fate selection.

    “PROX1, a signature gene defining granule neuron identity during pyramidal neuron fate selection44”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [44] La Manno, G. et al. RNA velocity of single cells. Nature 560, 494–498 (2018). source

      Not checked in the free run.

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

    Premisecitation

    ATAC/H3K27me3/H3K27ac Nano-CUT&Tag analysis of oligodendroglial differentiation in the juvenile brain showed two waves of H3K27me3 repression during this process.

    “These results are in accordance with ATAC/H3K27me3/H3K27ac Nano-CUT&Tag analysis37 of oligodendroglial differentiation in the juvenile brain, which showed two waves of H3K27me3 repression during this process37.”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [37] Bartosovic, M. & Castelo-Branco, G. Multimodal chromatin profiling using nanobody-based single-cell CUT&Tag. Nat. Biotechnol. https://doi.org/10.1038/s41587-022-01535-4 (2022). source

      Not checked in the free run.

  • Classified as probably not a substantial claim.

    Premisecitation

    For H3K27ac and H3K4me3, active genes should correspond to high gene activity score.

    “For H3K27ac and H3K4me3, active genes should correspond to high GAS2,13 (Methods).”
    Quote in the paperSubstantial claim0 of 2 sources checked
    • cites [2] Deng, Y. et al. Spatial-CUT&Tag: spatially resolved chromatin modification profiling at the cellular level. Science 375, 681–686 (2022). source

      Not checked in the free run.

    • cites [13] Granja, J. M. et al. ArchR is a scalable software package for integrative single-cell chromatin accessibility analysis. Nat. Genet. 53, 403–411 (2021). source

      Not checked in the free run.

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

    Study design and conductown data

    A biological replicate was performed on P21 mouse brain.

    “A biological replicate was also performed on P21 mouse brain (Extended Data Fig. 8; 20-μm pixel size, 50 × 50 barcodes).”
    Quote in the paperNot a substantial claim

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

    Premisecitation

    Ptprz1 is a marker of oligodendrocyte precursor cells.

    “such as Ptprz1, a marker of oligodendrocyte precursor cells38,39”
    Quote in the paperNot a substantial claim0 of 2 sources checked
    • cites [38] Marques, S. et al. Oligodendrocyte heterogeneity in the mouse juvenile and adult central nervous system. Science 352, 1326–1329 (2016). source

      Not checked in the free run.

    • cites [39] Marques, S. et al. Transcriptional convergence of oligodendrocyte lineage progenitors during development. Dev. Cell 46, 504–517 (2018). source

      Not checked in the free run.

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

    Interpretationreasoning

    The modest chromatin accessibility of PROX1 might be attributed to a minimal demand for synthesis of new PROX1 transcripts in postmitotic mature granule cells, and is thus not required to maintain an active open chromatin state.

    “This might be attributed to a minimal demand for synthesis of new PROX1 transcripts in postmitotic mature granule cells, and is thus not required to maintain an active open chromatin state.”
    Quote in the paperNot a substantial claim

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

    Study design and conductnone given

    All procedures described were approved by the local committee for ethical experiments on laboratory animals in Sweden.

    “All procedures described were approved by the local committee for ethical experiments on laboratory animals in Sweden (Stockholms Norra Djurförsöksetiska nämnd, nos. 1995/2019 and 7029/2020).”
    Quote in the paperNot a substantial claim

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

    Study design and conductnone given

    All experimental procedures were conducted following European directive no. 2010/63/EU, local Swedish directive no. L150/SJVFS/2019:9, Saknr no. L150 and Karolinska Institutet complementary guidelines for procurement and use of laboratory animals, no. Dnr. 1937/03-640.

    “All experimental procedures were conducted following European directive no. 2010/63/EU, local Swedish directive no. L150/SJVFS/2019:9, Saknr no. L150 and Karolinska Institutet complementary guidelines for procurement and use of laboratory animals, no. Dnr. 1937/03-640.”
    Quote in the paperNot a substantial claim

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

    Study design and conductnone given

    Human brain tissue collection was conducted with New York State Psychiatric Institute Institutional Review Board approval, and informed consent was obtained from next of kin.

    “Brain tissue collection was conducted with New York State Psychiatric Institute Institutional Review Board approval, and informed consent obtained from next of kin who agreed to donate the brains and participate in psychological autopsy interviews.”
    Quote in the paperNot a substantial claim
56 of 56

  • The paper relies on this without support.

    The selected tissues (E13 mouse embryo, P21/22 mouse brain, adult human hippocampus) are representative models for studying spatial epigenome–transcriptome relationships in mammalian tissues.

    “We applied these techniques to embryonic and juvenile mouse brain, as well as adult human brain hippocampus, to dissect epigenetic and transcriptional states and their role in the regulation of cell types dynamics in tissue.”

  • The paper relies on this without support.

    As of the time of the study, current technologies had not been able to perform unbiased genome-wide comapping of epigenome and transcriptome on the same tissue section at the cellular level.

    “As of today, current technologies have not been able to perform unbiased genome-wide comapping of epigenome and transcriptome on the same tissue section at the cellular level.”

  • The paper relies on this without support.

    To comprehensively understand the mechanism of gene regulation, different layers of omics information need to be profiled simultaneously.

    “However, to comprehensively understand the mechanism of gene regulation, different layers of omics information need to be profiled simultaneously.”
3 of 3

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?

  • Spatial ATAC-RNA-seq and spatial CUT&Tag-RNA-seq achieve co-profile of epigenome and transcriptome.

    “the group extended this approach to spatial ATAC-RNA-seq and spatial assay of cleavage under targets and tagmentation and RNA sequencing (spatial CUT&Tag-RNA-seq) [86], which achieve co-profile of epigenome and transcriptome”

    From Single-cell sequencing to multi-omics: technologies and applications (2024) · cites this paper as [86]

  • DBiT-seq has been extended to spatial co-profiling of mRNA and hundreds of proteins, and transcriptome and epigenome.

    “Recently, DBiT-seq has further demonstrated its versatility by spatial co-profiling of mRNA transcriptome and hundreds of cellular proteins as part of spatial CITE-seq126 and of transcriptome and epigenome as part of spatial ATAC–RNA-seq127, spatial CUT&Tag–RNA-seq127 and (reported in a preprint) SPARC-seq128.”

    From The technological landscape and applications of single-cell multi-omics (2023) · cites this paper as [127]

Cites

Flagged first, then by citations

Cited by

Sorted by citations

Related

Authors
Zhang, D., Deng, Y., Kukanja, P., Agirre, E., Bartosovic, M., Dong, M., Ma, C., Ma, S., Su, G., Bao, S., Liu, Y., Xiao, Y., Rosoklija, G. B., Dwork, A. J., Mann, J. J., Leong, K. W., Boldrini, M., Wang, L., Haeussler, M., Raphael, B. J., Kluger, Y., Castelo-Branco, G., & Fan, R.
Affiliations
  • Yale University (US)
  • Yale Cancer Center (US)
  • Karolinska Institutet (SE)
  • Princeton University (US)
  • Broad Institute (US)
  • Columbia University (US)
  • Macedonian Academy of Sciences and Arts (MK)
  • New York State Psychiatric Institute (US)
  • Columbia University Irving Medical Center (US)
  • University of California, Santa Cruz (US)
Corresponding author
  • Deng, Yanxiang
  • Fan, Rong
ORCID
Venue
Nature
Publisher
Springer Science and Business Media LLC
Published
2023-03-15
Volume
616
Issue
7955
Pages
113-122
ISSN
  • 0028-0836 (ISSN-L)
  • 1476-4687
DOI
10.1038/s41586-023-05795-1
PMC
PMC10076218
OpenAlex
W4324344255
Semantic Scholar
bde8da8ff6a896af73232add8cdb5f49a044ecc1
Cited by
483
References
58
Citations by year
2026: 172 · 2025: 163 · 2024: 87 · 2023: 59 · 2013: 1
FWCI
42.73
Type
journal article
Language
en
Open access
Licence
Locations
  • Nature (publishedVersion, cc-by)
  • Research Square (acceptedVersion, cc-by)
  • PubMed Central (publishedVersion, cc-by)
  • PubMed (publishedVersion)
  • RePEc: Research Papers in Economics (submittedVersion)
  • eScholarship (California Digital Library) (submittedVersion, cc-by)
Funders
  • Yale University
  • Li Ka Shing Foundation
  • Society for Immunotherapy of Cancer
  • Göran Gustafssons Stiftelser
  • European Commission (HORIZON2020, 794689)
  • Cancerfonden (190394 Pj, 2019-)
  • Svenska Sällskapet för Medicinsk Forskning (JUB2019)
  • Karolinska Institutet
  • Knut och Alice Wallenbergs Stiftelse (2019–0107)
  • Vetenskapsrådet (2019-01360)
  • National Institutes of Health (1u01ca294514-01, 1rf1mh128876-01)
  • National Institute on Aging (U54AG076043, U54AG079759)
  • National Institute of Mental Health (RF1 MH128876, RM1MH132648)
  • National Cancer Institute (U01 CA294514, UH3CA257393, UG3CA257393, U24CA248453, R01CA245313, U54CA274509)
  • National Institute of Diabetes and Digestive and Kidney Diseases (U54DK106857)
Topics
  • Life Sciences › Biochemistry, Genetics and Molecular Biology › Molecular Biology › Single-cell and spatial transcriptomics
  • Life Sciences › Biochemistry, Genetics and Molecular Biology › Molecular Biology › Genomics and Chromatin Dynamics
  • Life Sciences › Biochemistry, Genetics and Molecular Biology › Molecular Biology › Epigenetics and DNA Methylation
Keywords
spatial epigenomics, spatial transcriptomics, chromatin accessibility, histone modifications, H3K27me3, H3K27ac, H3K4me3, single-cell resolution, mouse brain, human brain, cell dynamics, gene expression regulation, epigenetic priming, tissue structure
MeSH terms
Epigenome, Aging, Animals, Brain, Chromatin, Gene Expression Regulation, Histones, Humans, Mammals, Organ Specificity, Gene Expression Profiling, Epigenesis, Genetic, Mice, Epigenomics, Single-Cell Analysis, Transcriptome

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 2026-10-07.

Classified by jev-1.13-20260917 (gold-guideline-v0.2) on 2026-10-07.

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