Genome-Scale Imaging of the 3D Organization and Transcriptional Activity of Chromatin

•Massively multiplexed FISH enables mapping chromatin structure at genome scale•Multimodal high-throughput imaging places chromatin structure in functional context•Trans-chromosome or long-range interactions occur preferentially among active chromatin•Transcription activity correlates with local enrichment of compartment A chromatin The 3D organization of chromatin regulates many genome functions. Our understanding of 3D genome organization requires tools to directly visualize chromatin conformation in its native context. Here we report an imaging technology for visualizing chromatin organization across multiple scales in single cells with high genomic throughput. First we demonstrate multiplexed imaging of hundreds of genomic loci by sequential hybridization, which allows high-resolution conformation tracing of whole chromosomes. Next we report a multiplexed error-robust fluorescence in situ hybridization (MERFISH)-based method for genome-scale chromatin tracing and demonstrate simultaneous imaging of more than 1,000 genomic loci and nascent transcripts of more than 1,000 genes together with landmark nuclear structures. Using this technology, we characterize chromatin domains, compartments, and trans-chromosomal interactions and their relationship to transcription in single cells. We envision broad application of this high-throughput, multi-scale, and multi-modal imaging technology, which provides an integrated view of chromatin organization in its native structural and functional context. The 3D organization of chromatin regulates many genome functions. Our understanding of 3D genome organization requires tools to directly visualize chromatin conformation in its native context. Here we report an imaging technology for visualizing chromatin organization across multiple scales in single cells with high genomic throughput. First we demonstrate multiplexed imaging of hundreds of genomic loci by sequential hybridization, which allows high-resolution conformation tracing of whole chromosomes. Next we report a multiplexed error-robust fluorescence in situ hybridization (MERFISH)-based method for genome-scale chromatin tracing and demonstrate simultaneous imaging of more than 1,000 genomic loci and nascent transcripts of more than 1,000 genes together with landmark nuclear structures. Using this technology, we characterize chromatin domains, compartments, and trans-chromosomal interactions and their relationship to transcription in single cells. We envision broad application of this high-throughput, multi-scale, and multi-modal imaging technology, which provides an integrated view of chromatin organization in its native structural and functional context. The 3D organization of the genome regulates many essential cellular functions ranging from gene expression to DNA replication (Cavalli and Heard, 2019Cavalli G. Heard E. Advances in epigenetics link genetics to the environment and disease.Nature. 2019; 571: 489-499Crossref PubMed Scopus (277) Google Scholar; Dekker and Mirny, 2016Dekker J. Mirny L. The 3D Genome as Moderator of Chromosomal Communication.Cell. 2016; 164: 1110-1121Abstract Full Text Full Text PDF PubMed Scopus (443) Google Scholar; Yu and Ren, 2017Yu M. Ren B. The Three-Dimensional Organization of Mammalian Genomes.Annu. Rev. Cell Dev. Biol. 2017; 33: 265-289Crossref PubMed Scopus (137) Google Scholar). Biochemical and imaging measurements have unveiled complex chromatin structures across a wide range of scales (Bickmore, 2013Bickmore W.A. The spatial organization of the human genome.Annu. Rev. Genomics Hum. Genet. 2013; 14: 67-84Crossref PubMed Scopus (227) Google Scholar; Cavalli and Heard, 2019Cavalli G. Heard E. Advances in epigenetics link genetics to the environment and disease.Nature. 2019; 571: 489-499Crossref PubMed Scopus (277) Google Scholar; Dekker and Mirny, 2016Dekker J. Mirny L. The 3D Genome as Moderator of Chromosomal Communication.Cell. 2016; 164: 1110-1121Abstract Full Text Full Text PDF PubMed Scopus (443) Google Scholar; Hnisz et al., 2017Hnisz D. Shrinivas K. Young R.A. Chakraborty A.K. Sharp P.A. A Phase Separation Model for Transcriptional Control.Cell. 2017; 169: 13-23Abstract Full Text Full Text PDF PubMed Scopus (653) Google Scholar; Krijger and de Laat, 2016Krijger P.H.L. de Laat W. Regulation of disease-associated gene expression in the 3D genome.Nat. Rev. Mol. Cell Biol. 2016; 17: 771-782Crossref PubMed Scopus (129) Google Scholar; Lakadamyali and Cosma, 2020Lakadamyali M. Cosma M.P. Visualizing the genome in high resolution challenges our textbook understanding.Nat. Methods. 2020; 17: 371-379Crossref PubMed Scopus (22) Google Scholar; Levine et al., 2014Levine M. Cattoglio C. Tjian R. Looping back to leap forward: transcription enters a new era.Cell. 2014; 157: 13-25Abstract Full Text Full Text PDF PubMed Scopus (289) Google Scholar; Yu and Ren, 2017Yu M. Ren B. The Three-Dimensional Organization of Mammalian Genomes.Annu. Rev. Cell Dev. Biol. 2017; 33: 265-289Crossref PubMed Scopus (137) Google Scholar). In particular, high-throughput chromosome conformation capture methods, such as Hi-C (Lieberman-Aiden et al., 2009Lieberman-Aiden E. van Berkum N.L. Williams L. Imakaev M. Ragoczy T. Telling A. Amit I. Lajoie B.R. Sabo P.J. Dorschner M.O. et al.Comprehensive Mapping of Long-Range Interactions Reveals Folding Principles of the Human Genome.Science. 2009; 326: 289-293Crossref PubMed Scopus (4135) Google Scholar), genome architecture mapping (GAM) (Beagrie et al., 2017Beagrie R.A. Scialdone A. Schueler M. Kraemer D.C.A. Chotalia M. Xie S.Q. Barbieri M. de Santiago I. Lavitas L.-M. Branco M.R. et al.Complex multi-enhancer contacts captured by genome architecture mapping.Nature. 2017; 543: 519-524Crossref PubMed Scopus (275) Google Scholar), and other sequencing-based methods (Dekker et al., 2017Dekker J. Belmont A.S. Guttman M. Leshyk V.O. Lis J.T. Lomvardas S. Mirny L.A. O’Shea C.C. Park P.J. Ren B. et al.4D Nucleome NetworkThe 4D nucleome project.Nature. 2017; 549: 219-226Crossref PubMed Scopus (238) Google Scholar; Yu and Ren, 2017Yu M. Ren B. The Three-Dimensional Organization of Mammalian Genomes.Annu. Rev. Cell Dev. Biol. 2017; 33: 265-289Crossref PubMed Scopus (137) Google Scholar), have revealed chromatin structures, such as domains and compartments, with a genome-wide view (Cavalli and Heard, 2019Cavalli G. Heard E. Advances in epigenetics link genetics to the environment and disease.Nature. 2019; 571: 489-499Crossref PubMed Scopus (277) Google Scholar; Dekker and Mirny, 2016Dekker J. Mirny L. The 3D Genome as Moderator of Chromosomal Communication.Cell. 2016; 164: 1110-1121Abstract Full Text Full Text PDF PubMed Scopus (443) Google Scholar; Krijger and de Laat, 2016Krijger P.H.L. de Laat W. Regulation of disease-associated gene expression in the 3D genome.Nat. Rev. Mol. Cell Biol. 2016; 17: 771-782Crossref PubMed Scopus (129) Google Scholar; Yu and Ren, 2017Yu M. Ren B. The Three-Dimensional Organization of Mammalian Genomes.Annu. Rev. Cell Dev. Biol. 2017; 33: 265-289Crossref PubMed Scopus (137) Google Scholar). Notably, chromatin is partitioned into genomic regions with enhanced self-interaction, called topologically associated domains (TADs) (Dixon et al., 2012Dixon J.R. Selvaraj S. Yue F. Kim A. Li Y. Shen Y. Hu M. Liu J.S. Ren B. Topological domains in mammalian genomes identified by analysis of chromatin interactions.Nature. 2012; 485: 376-380Crossref PubMed Scopus (3398) Google Scholar; Nora et al., 2012Nora E.P. Lajoie B.R. Schulz E.G. Giorgetti L. Okamoto I. Servant N. Piolot T. van Berkum N.L. Meisig J. Sedat J. et al.Spatial partitioning of the regulatory landscape of the X-inactivation centre.Nature. 2012; 485: 381-385Crossref PubMed Scopus (1559) Google Scholar; Sexton et al., 2012Sexton T. Yaffe E. Kenigsberg E. Bantignies F. Leblanc B. Hoichman M. Parrinello H. Tanay A. Cavalli G. Three-dimensional folding and functional organization principles of the Drosophila genome.Cell. 2012; 148: 458-472Abstract Full Text Full Text PDF PubMed Scopus (1147) Google Scholar), that often harbor co-regulated genes and have boundaries coinciding with regulatory epigenetic elements (Dekker and Mirny, 2016Dekker J. Mirny L. The 3D Genome as Moderator of Chromosomal Communication.Cell. 2016; 164: 1110-1121Abstract Full Text Full Text PDF PubMed Scopus (443) Google Scholar; Krijger and de Laat, 2016Krijger P.H.L. de Laat W. Regulation of disease-associated gene expression in the 3D genome.Nat. Rev. Mol. Cell Biol. 2016; 17: 771-782Crossref PubMed Scopus (129) Google Scholar; Yu and Ren, 2017Yu M. Ren B. The Three-Dimensional Organization of Mammalian Genomes.Annu. Rev. Cell Dev. Biol. 2017; 33: 265-289Crossref PubMed Scopus (137) Google Scholar). At a larger scale, chromatin is partitioned into A and B compartments that are enriched for active and inactive chromatin, respectively (Lieberman-Aiden et al., 2009Lieberman-Aiden E. van Berkum N.L. Williams L. Imakaev M. Ragoczy T. Telling A. Amit I. Lajoie B.R. Sabo P.J. Dorschner M.O. et al.Comprehensive Mapping of Long-Range Interactions Reveals Folding Principles of the Human Genome.Science. 2009; 326: 289-293Crossref PubMed Scopus (4135) Google Scholar). The physiological significance of A/B compartmentalization is implicated by its changes during development and between cell types (Yu and Ren, 2017Yu M. Ren B. The Three-Dimensional Organization of Mammalian Genomes.Annu. Rev. Cell Dev. Biol. 2017; 33: 265-289Crossref PubMed Scopus (137) Google Scholar). Although these high-throughput sequencing-based approaches are powerful and have greatly enriched our knowledge of 3D genome organization, they provide contact information for pairs of chromatin loci but not direct spatial position information of individual loci. Furthermore, most of the genome-wide insights into chromatin organization are built on population-averaged contact maps across millions of cells. Investigation of 3D genome organization in single cells remains a challenging task despite recent improvements in single-cell Hi-C methods (Flyamer et al., 2017Flyamer I.M. Gassler J. Imakaev M. Brandão H.B. Ulianov S.V. Abdennur N. Razin S.V. Mirny L.A. Tachibana-Konwalski K. Single-nucleus Hi-C reveals unique chromatin reorganization at oocyte-to-zygote transition.Nature. 2017; 544: 110-114Crossref PubMed Scopus (325) Google Scholar; Nagano et al., 2017Nagano T. Lubling Y. Várnai C. Dudley C. Leung W. Baran Y. Mendelson Cohen N. Wingett S. Fraser P. Tanay A. Cell-cycle dynamics of chromosomal organization at single-cell resolution.Nature. 2017; 547: 61-67Crossref PubMed Scopus (296) Google Scholar; Ramani et al., 2017Ramani V. Deng X. Qiu R. Gunderson K.L. Steemers F.J. Disteche C.M. Noble W.S. Duan Z. Shendure J. Massively multiplex single-cell Hi-C.Nat. Methods. 2017; 14: 263-266Crossref PubMed Scopus (231) Google Scholar; Stevens et al., 2017Stevens T.J. Lando D. Basu S. Atkinson L.P. Cao Y. Lee S.F. Leeb M. Wohlfahrt K.J. Boucher W. O’Shaughnessy-Kirwan A. et al.3D structures of individual mammalian genomes studied by single-cell Hi-C.Nature. 2017; 544: 59-64Crossref PubMed Scopus (361) Google Scholar; Tan et al., 2018Tan L. Xing D. Chang C.-H. Li H. Xie X.S. Three-dimensional genome structures of single diploid human cells.Science. 2018; 361: 924-928Crossref PubMed Scopus (142) Google Scholar). In addition, although methods have emerged to combine Hi-C with other measurement modalities—for example, to provide characterization of chromatin contacts in the context of interacting proteins (Fang et al., 2016Fang R. Yu M. Li G. Chee S. Liu T. Schmitt A.D. Ren B. Mapping of long-range chromatin interactions by proximity ligation-assisted ChIP-seq.Cell Res. 2016; 26: 1345-1348Crossref PubMed Scopus (121) Google Scholar; Fullwood et al., 2009Fullwood M.J. Liu M.H. Pan Y.F. Liu J. Xu H. Mohamed Y.B. Orlov Y.L. Velkov S. Ho A. Mei P.H. et al.An oestrogen-receptor-α-bound human chromatin interactome.Nature. 2009; 462: 58-64Crossref PubMed Scopus (1116) Google Scholar; Mumbach et al., 2016Mumbach M.R. Rubin A.J. Flynn R.A. Dai C. Khavari P.A. Greenleaf W.J. Chang H.Y. HiChIP: efficient and sensitive analysis of protein-directed genome architecture.Nat. Methods. 2016; 13: 919-922Crossref PubMed Scopus (385) Google Scholar; Zheng et al., 2019Zheng M. Tian S.Z. Capurso D. Kim M. Maurya R. Lee B. Piecuch E. Gong L. Zhu J.J. Li Z. et al.Multiplex chromatin interactions with single-molecule precision.Nature. 2019; 566: 558-562Crossref PubMed Scopus (79) Google Scholar), nuclear structures (Chen et al., 2018Chen Y. Zhang Y. Wang Y. Zhang L. Brinkman E.K. Adam S.A. Goldman R. van Steensel B. Ma J. Belmont A.S. Mapping 3D genome organization relative to nuclear compartments using TSA-Seq as a cytological ruler.J. Cell Biol. 2018; 217: 4025-4048Crossref PubMed Scopus (108) Google Scholar; Girelli et al., 2020Girelli G. Custodio J. Kallas T. Agostini F. Wernersson E. Spanjaard B. Mota A. Kolbeinsdottir S. Gelali E. Crosetto N. Bienko M. GPSeq reveals the radial organization of chromatin in the cell nucleus.Nat. Biotechnol. 2020; (Published online May 25, 2020)https://doi.org/10.1038/s41587-020-0519-yCrossref PubMed Scopus (13) Google Scholar; Quinodoz et al., 2018Quinodoz S.A. Ollikainen N. Tabak B. Palla A. Schmidt J.M. Detmar E. Lai M.M. Shishkin A.A. Bhat P. Takei Y. et al.Higher-Order Inter-chromosomal Hubs Shape 3D Genome Organization in the Nucleus.Cell. 2018; 174: 744-757.e24Abstract Full Text Full Text PDF PubMed Scopus (278) Google Scholar), or DNA modifications (Lee et al., 2019Lee D.-S. Luo C. Zhou J. Chandran S. Rivkin A. Bartlett A. Nery J.R. Fitzpatrick C. O’Connor C. Dixon J.R. Ecker J.R. Simultaneous profiling of 3D genome structure and DNA methylation in single human cells.Nat. Methods. 2019; 16: 999-1006Crossref PubMed Scopus (50) Google Scholar; Li et al., 2019Li G. Liu Y. Zhang Y. Kubo N. Yu M. Fang R. Kellis M. Ren B. Joint profiling of DNA methylation and chromatin architecture in single cells.Nat. Methods. 2019; 16: 991-993Crossref PubMed Scopus (51) Google Scholar)—multi-modal measurements by sequencing remain challenging. Notably, a method that allows genome-scale measurements of chromatin organization and transcriptional activity in the same cells has not emerged, despite the demand for such a method to further our understanding of how chromatin organization regulates transcription and vice versa. Imaging-based approaches, on the other hand, provide a direct measure of the spatial positions of chromatin loci in individual cells with high detection efficiency. In particular, fluorescence in situ hybridization (FISH) provides highly specific detection of chromatin loci in fixed cells (Levsky and Singer, 2003Levsky J.M. Singer R.H. Fluorescence in situ hybridization: past, present and future.J. Cell Sci. 2003; 116: 2833-2838Crossref PubMed Scopus (348) Google Scholar), and the clustered regularly interspaced short palindromic repeats (CRISPR) system substantially enhances our ability to image specific chromatin loci in live cells (Chen et al., 2016aChen B. Guan J. Huang B. Imaging Specific Genomic DNA in Living Cells.Annu. Rev. Biophys. 2016; 45: 1-23Crossref PubMed Scopus (39) Google Scholar; Knight et al., 2018Knight S.C. Tjian R. Doudna J.A. Genomes in Focus: Development and Applications of CRISPR-Cas9 Imaging Technologies.Angew. Chem. Int. Engl. 2018; 57: 4329-4337Crossref PubMed Scopus (34) Google Scholar; Xu and Qi, 2019Xu X. Qi L.S. A CRISPR-dCas Toolbox for Genetic Engineering and Synthetic Biology.J. Mol. Biol. 2019; 431: 34-47Crossref PubMed Scopus (97) Google Scholar). However, current imaging methods have limited throughput in sequence space, traditionally allowing the study of only a few genomic loci at a time. We recently developed a chromatin tracing approach by sequential rounds of FISH imaging that enabled imaging of tens of distinct chromatin loci in single cells (Wang et al., 2016Wang S. Su J.-H. Beliveau B.J. Bintu B. Moffitt J.R. Wu C. Zhuang X. Spatial organization of chromatin domains and compartments in single chromosomes.Science. 2016; 353: 598-602Crossref PubMed Scopus (262) Google Scholar). This approach provides insights into chromatin structures and their relationship with transcription (Bintu et al., 2018Bintu B. Mateo L.J. Su J. Sinnott-Armstrong N.A. Parker M. Kinrot S. Yamaya K. Boettiger A.N. Zhuang X. Super-resolution chromatin tracing reveals domains and cooperative interactions in single cells.Science. 2018; 362 (eaau1783)Crossref PubMed Scopus (316) Google Scholar; Cardozo Gizzi et al., 2019Cardozo Gizzi A.M. Cattoni D.I. Fiche J.-B. Espinola S.M. Gurgo J. Messina O. Houbron C. Ogiyama Y. Papadopoulos G.L. Cavalli G. et al.Microscopy-Based Chromosome Conformation Capture Enables Simultaneous Visualization of Genome Organization and Transcription in Intact Organisms.Mol. Cell. 2019; 74: 212-222.e5Abstract Full Text Full Text PDF PubMed Scopus (80) Google Scholar; Liu et al., 2020Liu M. Lu Y. Yang B. Chen Y. Radda J.S.D. Hu M. Katz S.G. Wang S. Multiplexed imaging of nucleome architectures in single cells of mammalian tissue.Nature Communications. 2020; 11: 2907Crossref PubMed Scopus (18) Google Scholar; Mateo et al., 2019Mateo L.J. Murphy S.E. Hafner A. Cinquini I.S. Walker C.A. Boettiger A.N. Visualizing DNA folding and RNA in embryos at single-cell resolution.Nature. 2019; 568: 49-54Crossref PubMed Scopus (128) Google Scholar; Nir et al., 2018Nir G. Farabella I. Pérez Estrada C. Ebeling C.G. Beliveau B.J. Sasaki H.M. Lee S.D. Nguyen S.C. McCole R.B. Chattoraj S. et al.Walking along chromosomes with super-resolution imaging, contact maps, and integrative modeling.PLoS Genet. 2018; 14: e1007872Crossref PubMed Scopus (99) Google Scholar; Sawh et al., 2020Sawh A.N. Shafer M.E.R. Su J. Zhuang X. Wang S. Mango S.E. Lamina-Dependent Stretching and Unconventional Chromosome Compartments in Early C. elegans Embryos.Molecular Cell. 2020; 78: 96-111Abstract Full Text Full Text PDF PubMed Scopus (17) Google Scholar; Wang et al., 2016Wang S. Su J.-H. Beliveau B.J. Bintu B. Moffitt J.R. Wu C. Zhuang X. Spatial organization of chromatin domains and compartments in single chromosomes.Science. 2016; 353: 598-602Crossref PubMed Scopus (262) Google Scholar). However, because the number of genomic loci that can be imaged in individual cells remains limited, a high-resolution view of whole chromosomes in single cells is still missing, not to mention a genome-scale view of chromatin organization in individual cells. Here we report a multi-scale multiplexed FISH imaging technology that enables simultaneous imaging of hundreds to more than 1,000 distinct genomic loci at various resolutions and genomic coverages in single cells. Using this approach, we further demonstrate simultaneous imaging of more than 1,000 genomic loci with nascent RNA transcripts of more than 1,000 genes residing in these loci and landmark nuclear structures, including nuclear speckles and nucleoli. This approach allows us to place chromatin organization in its native structural and functional context and, thus, to explore the relationship between chromatin organization, transcriptional activity, and nuclear structures in single cells. To enable a systematic view of chromatin structures across multiple scales, we developed an imaging platform (STAR Methods; Table S1) for direct visualization of chromatin with exceptionally high throughput in sequence space up to the genome scale. This platform includes two complementary approaches (Figure 1A). First, for imaging of chromatin structures that are relatively small so that different loci therein are difficult to resolve in any single image, we expanded our previously reported sequential imaging strategy (Wang et al., 2016Wang S. Su J.-H. Beliveau B.J. Bintu B. Moffitt J.R. Wu C. Zhuang X. Spatial organization of chromatin domains and compartments in single chromosomes.Science. 2016; 353: 598-602Crossref PubMed Scopus (262) Google Scholar) to allow tracing of hundreds of chromatin loci in single cells. In this approach, chromatin is imaged one locus at a time (or M loci at a time with M-color imaging) across many imaging rounds (Figure 1A, left). We demonstrated this approach by using it to trace the conformation of whole chromosomes in single cells at high resolution. Second, to image chromatin structures that occupy an area substantially larger than the diffraction-limited resolution, we developed a more efficient, combinatorial imaging strategy based on multiplexed error-robust FISH (MERFISH) (Chen et al., 2015Chen K.H. Boettiger A.N. Moffitt J.R. Wang S. Zhuang X. Spatially resolved, highly multiplexed RNA profiling in single cells.Science. 2015; 348: aaa6090Crossref PubMed Scopus (694) Google Scholar), where many chromatin loci are imaged simultaneously in each round, and their distinct identities are determined based on the combinations of rounds in which they appear (Figure 1A, right). The latter approach allows more genomic loci to be imaged in fewer imaging rounds. We used this approach to provide a genome-scale view of chromatin organization in the context of transcriptional activity and important nuclear structures in single cells. In this section, we describe high-resolution whole-chromosome tracing using the sequential imaging approach (Figure 1A, left; Figure S1A). We first focused on human chromosome 21 (Chr21) and partitioned the non-repetitive portion of the chromosome (Chr21, 10.4–46.7 Mb, from the hg38 assembly) into ~650 segments (i.e., ~650 genomic loci), each 50 kb in length (see STAR Methods for details). We then designed a library of primary oligonucleotide probes (Table S2), each containing a variable target sequence for hybridizing to the different regions of the chromosome and a readout sequence that is unique to each of the 50-kb loci and can be detected by complementary fluorescent readout probes (Figure S1A). However, identifying these loci with ~650 distinct, dye-labeled readout probes would be prohibitively expensive, given the high cost of dye-labeled oligonucleotides. To overcome this challenge, we devised a two-step labeling strategy to detect the distinct readout sequences with a common set of three dye-labeled readout probes (one for each color channel) mediated by unlabeled adaptor probes that convert each locus-specific readout sequence into one of the three common readout sequences (Figure S1A; Table S3). Using this strategy, we sequentially imaged ~650 genomic loci in Chr21 in human lung fibroblast (IMR-90) cells using more than 200 rounds of hybridization with three-color imaging. To enable stable imaging over such a large number of rounds, we optimized our imaging protocol to (1) maintain sample integrity and primary probe binding stability and (2) ensure complete removal of the fluorescence signal after each imaging round and minimize the accumulation of residual signal across hundreds of labeling rounds (see STAR Methods for a detailed experimental protocol). After imaging, we determined the centroid position of each chromatin locus in 3D and reconstructed the conformation of each copy of Chr21 in each cell (Figure 1B). To estimate how stable the sample and imaging instruments were over many hybridization rounds, we re-imaged a subset of loci after different numbers of rounds and used the displacement between the originally imaged locations and re-imaged locations of the loci to assess measurement accuracy. The median displacement increased from ~70 nm to ~120 nm when the number of hybridization rounds separating the original and re-imaging instances increased from ~10 to ~250 (Figures S1B and S1C). Loci that displayed greater displacement upon re-imaging tended to have a lower fluorescence signal intensity (Figure S1D). Among the loci that had large spatial distances from neighboring loci, a fraction of them also exhibited large re-imaging displacement errors and low brightness, suggesting relatively low confidence in the localizations of these loci that are far away from both of their genomic neighbors (see STAR Methods for details). Overall, the median displacement errors for individual loci upon re-imaging were substantially smaller than the median distances between neighboring chromatin loci (~250 nm) (Figure S1B). In addition, median pairwise distances between imaged loci were highly reproducible between biological replicates (Figure S1E). The detection efficiency of the chromatin loci in these experiments was more than 90%. To obtain a population-averaged view of the conformation of Chr21, we quantified the pairwise interaction between imaged loci by calculating their median spatial distance and their proximity frequency (i.e., the probability that the loci come into proximity) across all imaged chromosome copies (~12,100 copies from the two replicates) (Figure 1C; Figures S1F–S1I). We observed a high correlation between our median pairwise distances from the imaging data and previously published Hi-C data (Figures S1F–S1I). To choose the cutoff distance below which we consider two loci to be in proximity, we calculated the Pearson correlation coefficient between the Hi-C data (Rao et al., 2014Rao S.S.P. Huntley M.H. Durand N.C. Stamenova E.K. Bochkov I.D. Robinson J.T. Sanborn A.L. Machol I. Omer A.D. Lander E.S. Aiden E.L. A 3D map of the human genome at kilobase resolution reveals principles of chromatin looping.Cell. 2014; 159: 1665-1680Abstract Full Text Full Text PDF PubMed Scopus (2888) Google Scholar) and the proximity frequencies derived from our imaging data across a range of cutoff distances. The Pearson correlation coefficient remained high for a wide range of cutoff distances but peaked at 0.87 when the cutoff distance was ~400–600 nm (Figure S1J). We thus chose 500 nm as the cutoff distance to generate proximity frequency maps throughout this work. The median distance and the proximity frequency maps showed block-like TAD structures (Figure 1C; Figures S1F and S1H). TAD boundaries identified from the distance and proximity frequency maps from our imaging data were highly similar to those determined from Hi-C data and were not sensitive to the localization error (~100 nm) in our chromatin traces (Figure S1K). Chromosomes were partitioned into domains that manifested as on-diagonal block-like features in single-cell spatial distance matrices, which also exhibited complex off-diagonal features indicating longer-range interactions (Figure S1L). These domains, long-range interactions, and the inter-locus distances showed high cell-to-cell variability (Figures S1L and S1M), consistent with the substantial variability in chromatin contacts observed in single-cell Hi-C data (Flyamer et al., 2017Flyamer I.M. Gassler J. Imakaev M. Brandão H.B. Ulianov S.V. Abdennur N. Razin S.V. Mirny L.A. Tachibana-Konwalski K. Single-nucleus Hi-C reveals unique chromatin reorganization at oocyte-to-zygote transition.Nature. 2017; 544: 110-114Crossref PubMed Scopus (325) Google Scholar; Tan et al., 2018Tan L. Xing D. Chang C.-H. Li H. Xie X.S. Three-dimensional genome structures of single diploid human cells.Science. 2018; 361: 924-928Crossref PubMed Scopus (142) Google Scholar). We previously observed similar single-cell domains and cell-to-cell variability when imaging small (~2-Mb) regions of the chromosome (Bintu et al., 2018Bintu B. Mateo L.J. Su J. Sinnott-Armstrong N.A. Parker M. Kinrot S. Yamaya K. Boettiger A.N. Zhuang X. Super-resolution chromatin tracing reveals domains and cooperative interactions in single cells.Science. 2018; 362 (eaau1783)Crossref PubMed Scopus (316) Google Scholar). However, within these small regions, a sizable fraction of cells did not display clear domain boundaries, and it remained unclear whether domains did not form within those cells or whether the entire imaged regions were within a single domain. Furthermore, the small size of the imaged regions prohibited accurate characterization of certain domain properties, such as their physical and genomic sizes. The high genomic throughput in this study provides a whole-chromosome view of these single-cell domain structures, revealing their presence in most of the imaged chromosome copies, allowing us to characterize their properties in a more systematic manner. We first identified the genomic locations of the boundaries of these single-cell domains. Although we observed a non-zero probability of boundary formation at nearly all imaged genomic loci, the domain boundaries were preferentially positioned near the binding sites of CTCF and cohesin (Figures 1C and 1D). In addition to the cell-to-cell variations in domain boundary positions, we also observed substantial heterogeneity in other features of these single-cell domains, ranging from their physical size to the degree of insulation or interaction between domains (Figures 1E–1H). Single-cell domains were variable in their genomic size (Figure 1I) and physical size (Figures 1E and 1J), neither of which was sensitive to our estimated locus localization error (Figures 1I and 1J). Domains bounded by the same genomic regions or having the same genomic size varied considerably in physical size from cell to cell (Figure 1E; Figure S1N). Interestingly, domains bounded by pairs of CTCF/cohesin binding sites that form loops in Hi-C maps (Rao et al., 2014Rao S.S.P. Huntley M.H. Durand N.C. Stamenova E.K. Bochkov I.D. Robinson J.T. Sanborn A.L. Machol I. Omer A.D. Lander E.S. Aiden E.L. A 3D map of the human genome at kilobase resolution reveals principles of chromatin looping.Cell. 2014; 159: 1665-1680Abstract Full Text Full Text PDF PubMed Scopus (2888) Google Scholar) tended to be smaller in physical size than domains not bounded by such genomic loci (Figure 1K). In addition, the degree of physical segregation between neighboring domains also varied substantially (Figures 1F and 1L), with some neighboring domains completely segregated in space w