High-resolution Hi-C Research Articles

HiCDiffusion - diffusion-enhanced, transformer-based prediction of chromatin interactions from DNA sequences

Prediction of chromatin interactions from DNA sequence has been a significant research challenge in the last couple of years. Several solutions have been proposed, most of which are based on encoder-decoder architecture, where 1D sequence is convoluted, encoded into the latent representation, and then decoded using 2D convolutions into the Hi-C pairwise chromatin spatial proximity matrix. Those methods, while obtaining high correlation scores and improved metrics, produce Hi-C matrices that are artificial - they are blurred due to the deep learning model architecture. In our study, we propose the HiCDiffusion, sequence-only model that addresses this problem. We first train the encoder-decoder neural network and then use it as a component of the diffusion model - where we guide the diffusion using a latent representation of the sequence, as well as the final output from the encoder-decoder. That way, we obtain the high-resolution Hi-C matrices that not only better resemble the experimental results - improving the Fréchet inception distance by an average of 11 times, with the highest improvement of 56 times - but also obtain similar classic metrics to current state-of-the-art encoder-decoder architectures used for the task.

CRISPR Screening of Transcribed Super-Enhancers Identifies Drivers of Triple-Negative Breast Cancer Progression.

Triple-negative breast cancer (TNBC) is the most therapeutically recalcitrant form of breast cancer, which is due in part to the paucity of targeted therapies. A systematic analysis of regulatory elements that extend beyond protein coding genes could uncover avenues for therapeutic intervention. To this end, we analyzed the regulatory mechanisms of TNBC-specific transcriptional enhancers together with their non-coding enhancer RNA (eRNA) transcripts. The functions of the top 30 eRNA-producing super-enhancers were systematically probed using high-throughput CRISPR-interference assays coupled to RNA-seq that enabled unbiased detection of target genes genome-wide. Generation of high resolution Hi-C chromatin interaction maps enabled annotation of the direct target genes for each super-enhancer, which highlighted their proclivity for genes that portend worse clinical outcomes in TNBC patients. Illustrating the utility of this dataset, deletion of an identified super-enhancer controlling the nearby PODXL gene or specific degradation of its enhancer RNAs led to profound inhibitory effects on target gene expression, cell proliferation, and migration. Furthermore, loss of this super-enhancer suppressed tumor growth and metastasis in TNBC mouse xenograft models. Single-cell RNA-seq and ATAC-seq analyses demonstrated the enhanced activity of this super-enhancer within the malignant cells of TNBC tumor specimens compared to non-malignant cell types. Collectively, this work examines several fundamental questions about how regulatory information encoded into eRNA-producing super-enhancers drives gene expression networks that underlie the biology of triple-negative breast cancer.

Enhancing Hi-C contact matrices for loop detection with Capricorn: a multiview diffusion model.

High-resolution Hi-C contact matrices reveal the detailed three-dimensional architecture of the genome, but high-coverage experimental Hi-C data are expensive to generate. Simultaneously, chromatin structure analyses struggle with extremely sparse contact matrices. To address this problem, computational methods to enhance low-coverage contact matrices have been developed, but existing methods are largely based on resolution enhancement methods for natural images and hence often employ models that do not distinguish between biologically meaningful contacts, such as loops and other stochastic contacts. We present Capricorn, a machine learning model for Hi-C resolution enhancement that incorporates small-scale chromatin features as additional views of the input Hi-C contact matrix and leverages a diffusion probability model backbone to generate a high-coverage matrix. We show that Capricorn outperforms the state of the art in a cross-cell-line setting, improving on existing methods by 17% in mean squared error and 26% in F1 score for chromatin loop identification from the generated high-coverage data. We also demonstrate that Capricorn performs well in the cross-chromosome setting and cross-chromosome, cross-cell-line setting, improving the downstream loop F1 score by 14% relative to existing methods. We further show that our multiview idea can also be used to improve several existing methods, HiCARN and HiCNN, indicating the wide applicability of this approach. Finally, we use DNA sequence to validate discovered loops and find that the fraction of CTCF-supported loops from Capricorn is similar to those identified from the high-coverage data. Capricorn is a powerful Hi-C resolution enhancement method that enables scientists to find chromatin features that cannot be identified in the low-coverage contact matrix. Implementation of Capricorn and source code for reproducing all figures in this paper are available at https://github.com/CHNFTQ/Capricorn.

Stratifying TAD boundaries pinpoints focal genomic regions of regulation, damage, and repair.

Advances in chromatin mapping have exposed the complex chromatin hierarchical organization in mammals, including topologically associating domains (TADs) and their substructures, yet the functional implications of this hierarchy in gene regulation and disease progression are not fully elucidated. Our study delves into the phenomenon of shared TAD boundaries, which are pivotal in maintaining the hierarchical chromatin structure and regulating gene activity. By integrating high-resolution Hi-C data, chromatin accessibility, and DNA double-strand breaks (DSBs) data from various cell lines, we systematically explore the complex regulatory landscape at high-level TAD boundaries. Our findings indicate that these boundaries are not only key architectural elements but also vibrant hubs, enriched with functionally crucial genes and complex transcription factor binding site-clustered regions. Moreover, they exhibit a pronounced enrichment of DSBs, suggesting a nuanced interplay between transcriptional regulation and genomic stability. Our research provides novel insights into the intricate relationship between the 3D genome structure, gene regulation, and DNA repair mechanisms, highlighting the role of shared TAD boundaries in maintaining genomic integrity and resilience against perturbations. The implications of our findings extend to understanding the complexities of genomic diseases and open new avenues for therapeutic interventions targeting the structural and functional integrity of TAD boundaries.

Light control ofthree-dimensional chromatin organization in soybean.

Higher-order chromatin structure is critical for regulation of gene expression. In plants, light profoundly affects the morphogenesis of emerging seedlings as well as global gene expression to ensure optimal adaptation to environmental conditions. However, the changes and functional significance of chromatin organization in response to light during seedling development are not well documented. We constructed Hi-C contact maps for the cotyledon, apical hook and hypocotyl of soybean subjected to dark and light conditions. The resulting high-resolution Hi-C contact maps identified chromosome territories, A/B compartments, A/B sub-compartments, TADs (Topologically Associated Domains) and chromatin loops in each organ. We observed increased chromatin compaction under light and we found that domains that switched from B sub-compartments in darkness to A sub-compartments under light contained genes that were activated during photomorphogenesis. At the local scale, we identified a group of TADs constructed by gene clusters consisting of different numbers of Small Auxin-Upregulated RNAs (SAURs), which exhibited strict co-expression in the hook and hypocotyl in response to light stimulation. In the hypocotyl, RNA polymerase II (RNAPII) regulated the transcription of a SAURs cluster under light via TAD condensation. Our results suggest that the 3D genome is involved in the regulation of light-related gene expression in a tissue-specific manner.

Learning Micro-C from Hi-C with diffusion models.

In the last few years, Micro-C has shown itself as an improved alternative to Hi-C. It replaced the restriction enzymes in Hi-C assays with micrococcal nuclease (MNase), resulting in capturing nucleosome resolution chromatin interactions. The signal-to-noise improvement of Micro-C allows it to detect more chromatin loops than high-resolution Hi-C. However, compared with massive Hi-C datasets available in the literature, there are only a limited number of Micro-C datasets. To take full advantage of these Hi-C datasets, we present HiC2MicroC, a computational method learning and then predicting Micro-C from Hi-C based on the denoising diffusion probabilistic models (DDPM). We trained our DDPM and other regression models in human foreskin fibroblast (HFFc6) cell line and evaluated these methods in six different cell types at 5-kb and 1-kb resolution. Our evaluations demonstrate that both HiC2MicroC and regression methods can markedly improve Hi-C towards Micro-C, and our DDPM-based HiC2MicroC outperforms regression in various terms. First, HiC2MicroC successfully recovers most of the Micro-C loops even those not detected in Hi-C maps. Second, a majority of the HiC2MicroC-recovered loops anchor CTCF binding sites in a convergent orientation. Third, HiC2MicroC loops share genomic and epigenetic properties with Micro-C loops, including linking promoters and enhancers, and their anchors are enriched for structural proteins (CTCF and cohesin) and histone modifications. Lastly, we find our recovered loops are also consistent with the loops identified from promoter capture Micro-C (PCMicro-C) and Chromatin Interaction Analysis by Paired-End Tag Sequencing (ChIA-PET). Overall, HiC2MicroC is an effective tool for further studying Hi-C data with Micro-C as a template. HiC2MicroC is publicly available at https://github.com/zwang-bioinformatics/HiC2MicroC/.

Cooltools: Enabling high-resolution Hi-C analysis in Python.

Chromosome conformation capture (3C) technologies reveal the incredible complexity of genome organization. Maps of increasing size, depth, and resolution are now used to probe genome architecture across cell states, types, and organisms. Larger datasets add challenges at each step of computational analysis, from storage and memory constraints to researchers' time; however, analysis tools that meet these increased resource demands have not kept pace. Furthermore, existing tools offer limited support for customizing analysis for specific use cases or new biology. Here we introduce cooltools (https://github.com/open2c/cooltools), a suite of computational tools that enables flexible, scalable, and reproducible analysis of high-resolution contact frequency data. Cooltools leverages the widely-adopted cooler format which handles storage and access for high-resolution datasets. Cooltools provides a paired command line interface (CLI) and Python application programming interface (API), which respectively facilitate workflows on high-performance computing clusters and in interactive analysis environments. In short, cooltools enables the effective use of the latest and largest genome folding datasets.

Building Haplotype-Resolved 3D Genome Maps of Chicken Skeletal Muscle.

Haplotype-resolved 3D chromatin architecture related to allelic differences in avian skeletal muscle development has not been addressed so far, although chicken husbandry for meat consumption has been prevalent feature of cultures on every continent for more than thousands of years. Here, high-resolution Hi-C diploid maps (1.2-kb maximum resolution) are generated for skeletal muscle tissues in chicken across three developmental stages (embryonic day 15 to day 30 post-hatching). The sequence features governing spatial arrangement of chromosomes and characterize homolog pairing in the nucleus, are identified. Multi-scale characterization of chromatin reorganization between stages from myogenesis in the fetus to myofiber hypertrophy after hatching show concordant changes in transcriptional regulation by relevant signaling pathways. Further interrogation of parent-of-origin-specific chromatin conformation supported that genomic imprinting is absent in birds. This study also reveals promoter-enhancer interaction (PEI) differences between broiler and layer haplotypes in skeletal muscle development-related genes are related to genetic variation between breeds, however, only a minority of breed-specific variations likely contribute to phenotypic divergence in skeletal muscle potentially via allelic PEI rewiring. Beyond defining the haplotype-specific 3D chromatin architecture in chicken, this study provides a rich resource for investigating allelic regulatory divergence among chicken breeds.

SpecLoop predicts cell type-specific chromatin loop via transcription factor cooperation

Cell-type-Specific Chromatin Loops (CSCLs) are crucial for gene regulation and cell fate determination. However, the mechanisms governing their establishment remain elusive. Here, we present SpecLoop, a network regularization-based machine learning framework, to investigate the role of transcription factors (TFs) cooperation in CSCL formation. SpecLoop integrates multi-omics data, including gene expression, chromatin accessibility, sequence, protein-protein interaction, and TF binding motif data, to predict CSCLs and identify TF cooperations. Using high resolution Hi-C data as the gold standard, SpecLoop accurately predicts CSCL in GM12878, IMR90, HeLa–S3, K562, HUVEC, HMEC, and NHEK seven cell types, with the AUROC values ranging from 0.8645 to 0.9852 and AUPR values ranging from 0.8654 to 0.9734. Notably SpecLoop demonstrates improved accuracy in predicting long-distance CSCLs and identifies TF complexes with strong predictive ability. Our study systematically explores the TFs and TF pairs associated with CSCL through effective integration of diverse omics data. SpecLoop is freely available at https://github.com/AMSSwanglab/SpecLoop.

MAE-seq refines regulatory elements across the genome.

Proper cell fate determination relies on precise spatial and temporal genome-wide cooperation between regulatory elements (REs) and their targeted genes. However, the lengths of REs defined using different methods vary, which indicates that there is sequence redundancy and that the context of the genome may be unintelligible. We developed a method called MAE-seq (Massive Active Enhancers by Sequencing) to experimentally identify functional REs at a 25-bp scale. In this study, MAE-seq was used to identify 626879, 541617 and 554826 25-bp enhancers in mouse embryonic stem cells (mESCs), C2C12 and HEK 293T, respectively. Using ∼1.6 trillion 25 bp DNA fragments and screening 12 billion cells, we identified 626879 as active enhancers in mESCs as an example. Comparative analysis revealed that most of the histone modification datasets were annotated by MAE-Seq loci. Furthermore, 33.85% (212195) of the identified enhancers were identified as de novo ones with no epigenetic modification. Intriguingly, distinct chromatin states dictate the requirement for dissimilar cofactors in governing novel and known enhancers. Validation results show that these 25-bp sequences could act as a functional unit, which shows identical or similar expression patterns as the previously defined larger elements, Enhanced resolution facilitated the identification of numerous cell-specific enhancers and their accurate annotation as super enhancers. Moreover, we characterized novel elements capable of augmenting gene activity. By integrating with high-resolution Hi-C data, over 55.64% of novel elements may have a distal association with different targeted genes. For example, we found that the Cdh1 gene interacts with one novel and two known REs in mESCs. The biological effects of these interactions were investigated using CRISPR-Cas9, revealing their role in coordinating Cdh1 gene expression and mESC proliferation. Our study presents an experimental approach to refine the REs at 25-bp resolution, advancing the precision of genome annotation and unveiling the underlying genome context. This novel approach not only advances our understanding of gene regulation but also opens avenues for comprehensive exploration of the genomic landscape.

Stage-specific dynamic reorganization of genome topology shapes transcriptional neighborhoods in developing human retinal organoids

We have generated a high-resolution Hi-C map of developing human retinal organoids to elucidate spatiotemporal dynamics of genomic architecture and its relationship with gene expression patterns. We demonstrate progressive stage-specific alterations in DNA topology and correlate these changes with transcription of cell-type-restricted gene markers during retinal differentiation. Temporal Hi-C reveals a shift toward A compartment for protein-coding genes and B compartment for non-coding RNAs, displaying high and low expression, respectively. Notably, retina-enriched genes are clustered near lost boundaries of topologically associated domains (TADs), and higher-order assemblages (i.e., TAD cliques) localize in active chromatin regions with binding sites for eye-field transcription factors. These genes gain chromatin contacts at their transcription start site as organoid differentiation proceeds. Our study provides a global view of chromatin architecture dynamics associated with diversification of cell types during retinal development and serves as a foundational resource for in-depth functional investigations of retinal developmental traits.

EPCO-17. 3D GENOME STRUCTURE REGULATES TRANSCRIPTION IN PEDIATRIC HIGH-GRADE GLIOMA

Abstract INTRODUCTION Pediatric high-grade gliomas (pHGGs), including glioblastoma multiforme (GBM) and diffuse intrinsic pontine glioma (DIPG), are highly morbid brain tumors. Up to 80% of DIPGs harbor a somatic missense mutation in genes encoding Histone H3. To investigate whether the H3K27M mutant protein is associated with distinct chromatin structure affecting transcription regulation, we generated the first high-resolution Hi-C and ATAC-Seq maps of pHGG cell lines, and integrated these with tissue and cell genomic data. METHODS We generated sequencing data from patient-derived cell lines (DIPG n=6, GBM n=3, normal n=2) and frozen tissue specimens (DIPG n=1, normal brainstem n=1). Analyses included cell line RNA-Seq, ChIP-Seq (H3K27ac, H3K27me3, H3K27M) and genome-wide chromatin conformation capture (Hi-C), as well as tissue ATAC-Seq. Publicly available pediatric glioma tissue ChIP-Seq data was integrated with cell data. CRISPR knock-down of target enhancer regions was performed. RESULTS We identified tumor-specific enhancers and regulatory networks for known oncogenes in DIPG and GBM. In DIPG, FOX, SOX, STAT and SMAD families were among top H3K27Ac enriched motifs. Significant differences in Topologically Associating Domains (TADs) and DNA looping were observed at OLIG2 and MYCN in H3K27M mutant DIPG, relative to wild-type GBM and normal cells. Pharmacologic treatment targeting H3K27Ac (BET and Bromodomain inhibition) altered these 3D structures. Functional analysis of differentially enriched enhancers in DIPG implicated SOX2, SUZ12, and TRIM24 as top activated upstream regulators. Distinct genomic structural variations leading to enhancer hijacking and gene co-amplification were identified at A2M, JAG2, and FLRT1. CONCLUSION We show genome structural variations enhancer-promoter interactions that impact gene expression in pHGG in the presence and absence of the H3K27M mutation. Our results imply that tridimensional genome alterations may play a critical role in the pHGG epigenetic landscape and thereby contribute to pediatric gliomagenesis. Further studies examining the impact of the alterations is therefore underway.

A 3D Chromosome Structure Reconstruction method with High Resolution Hi-C Data using Nonlinear Dimensionality Reduction and Divide-and-conquer strategy.

Chromosomes are fundamental components of genetic material, and their structural characteristics play an essential role in the regulation of gene expression. The advent of high-resolution Hi-C data has enabled scientists to explore the three-dimensional structure of chromosomes. However, most of the currently available methods for reconstructing chromosome structures are unable to achieve high resolutions, such as 5 Kilobase (KB). In this study, we present NeRV-3D, an innovative method that utilizes a nonlinear dimensionality reduction visualization algorithm to reconstruct 3D chromosome structures at low resolutions. Additionally, we introduce NeRV-3D-DC, which employs a divide-and-conquer technique to reconstruct and visualize 3D chromosome structures at high resolutions. Our results demonstrate that both NeRV-3D and NeRV-3D-DC outperform existing methods in terms of 3D visualization effects and evaluation metrics on simulated and actual Hi-C datasets. The implementation of NeRV-3D-DC can be found at https://github.com/ghaiyan/NeRV-3D-DC.

DFHiC: a dilated full convolution model to enhance the resolution of Hi-C data.

Hi-C technology has been the most widely used chromosome conformation capture (3C) experiment that measures the frequency of all paired interactions in the entire genome, which is a powerful tool for studying the 3D structure of the genome. The fineness of the constructed genome structure depends on the resolution of Hi-C data. However, due to the fact that high-resolution Hi-C data require deep sequencing and thus high experimental cost, most available Hi-C data are in low-resolution. Hence, it is essential to enhance the quality of Hi-C data by developing the effective computational methods. In this work, we propose a novel method, so-called DFHiC, which generates the high-resolution Hi-C matrix from the low-resolution Hi-C matrix in the framework of the dilated convolutional neural network. The dilated convolution is able to effectively explore the global patterns in the overall Hi-C matrix by taking advantage of the information of the Hi-C matrix in a way of the longer genomic distance. Consequently, DFHiC can improve the resolution of the Hi-C matrix reliably and accurately. More importantly, the super-resolution Hi-C data enhanced by DFHiC is more in line with the real high-resolution Hi-C data than those done by the other existing methods, in terms of both chromatin significant interactions and identifying topologically associating domains. https://github.com/BinWangCSU/DFHiC.

3D chromatin maps of the human pancreas reveal lineage-specific regulatory architecture of T2D risk

Three-dimensional (3D) chromatin organization maps help dissect cell-type-specific gene regulatory programs. Furthermore, 3D chromatin maps contribute to elucidating the pathogenesis of complex genetic diseases by connecting distal regulatory regions and genetic risk variants to their respective target genes. To understand the cell-type-specific regulatory architecture of diabetes risk, we generated transcriptomic and 3D epigenomic profiles of human pancreatic acinar, alpha, and beta cells using single-cell RNA-seq, single-cell ATAC-seq, and high-resolution Hi-C of sorted cells. Comparisons of these profiles revealed differential A/B (open/closed) chromatin compartmentalization, chromatin looping, and transcriptional factor-mediated control of cell-type-specific gene regulatory programs. We identified a total of 4,750 putative causal-variant-to-target-gene pairs at 194 type 2 diabetes GWAS signals using pancreatic 3D chromatin maps. We found that the connections between candidate causal variants and their putative target effector genes are cell-type stratified and emphasize previously underappreciated roles for alpha and acinar cells in diabetes pathogenesis.

The 3D architecture of the pepper genome and its relationship to function and evolution

The organization of chromatin into self-interacting domains is universal among eukaryotic genomes, though how and why they form varies considerably. Here we report a chromosome-scale reference genome assembly of pepper (Capsicum annuum) and explore its 3D organization through integrating high-resolution Hi-C maps with epigenomic, transcriptomic, and genetic variation data. Chromatin folding domains in pepper are as prominent as TADs in mammals but exhibit unique characteristics. They tend to coincide with heterochromatic regions enriched with retrotransposons and are frequently embedded in loops, which may correlate with transcription factories. Their boundaries are hotspots for chromosome rearrangements but are otherwise depleted for genetic variation. While chromatin conformation broadly affects transcription variance, it does not predict differential gene expression between tissues. Our results suggest that pepper genome organization is explained by a model of heterochromatin-driven folding promoted by transcription factories and that such spatial architecture is under structural and functional constraints.

HGG-02. Epigenetic transcription regulation and 3D genome structure in pediatric high-grade glioma

INTRODUCTION: Pediatric high-grade gliomas (pHGGs), including glioblastoma multiforme (GBM) and diffuse intrinsic pontine glioma (DIPG), are highly morbid brain tumors. Up to 80% of DIPGs harbor a somatic missense mutation in genes encoding Histone H3. To investigate whether the H3K27M mutant protein is associated with distinct chromatin structure affecting transcription regulation, we generated the first high-resolution Hi-C and ATAC-Seq maps of pHGG cell lines, and integrated these with tissue and cell genomic data. METHODS: We generated sequencing data from patient-derived cell lines (DIPG n=6, GBM n=3, normal n=2) and frozen tissue specimens (DIPG n=1, normal brainstem n=1). Analyses included cell line RNA-Seq, ChIP-Seq (H3K27ac, H3K27me3, H3K27M) and genome-wide chromatin conformation capture (Hi-C), as well as tissue ATAC-Seq. Publicly available pediatric glioma tissue ChIP-Seq data was integrated with cell data. CRISPR knock-down of target enhancer regions was performed. RESULTS: We identified tumor-specific enhancers and regulatory networks for known oncogenes in DIPG and GBM. In DIPG, FOX, SOX, STAT and SMAD families were among top H3K27Ac enriched motifs. Significant differences in Topologically Associating Domains (TADs) and DNA looping were observed at OLIG2 and MYCN in H3K27M mutant DIPG, relative to wild-type GBM and normal cells. Pharmacologic treatment targeting H3K27Ac (BET and Bromodomain inhibition) altered these 3D structures. Functional analysis of differentially enriched enhancers in DIPG implicated SOX2, SUZ12, and TRIM24 as top activated upstream regulators. Distinct genomic structural variations leading to enhancer hijacking and gene co-amplification were identified at A2M, JAG2, and FLRT1. CONCLUSION: We show genome structural variations enhancer-promoter interactions that impact gene expression in pHGG in the presence and absence of the H3K27M mutation. Our results imply that tridimensional genome alterations may play a critical role in the pHGG epigenetic landscape and thereby contribute to pediatric gliomagenesis. Further studies examining the impact of the alterations are therefore underway.

Comparative 3D genome architecture in vertebrates.

BackgroundThe three-dimensional (3D) architecture of the genome has a highly ordered and hierarchical nature, which influences the regulation of essential nuclear processes at the basis of gene expression, such as gene transcription. While the hierarchical organization of heterochromatin and euchromatin can underlie differences in gene expression that determine evolutionary differences among species, the way 3D genome architecture is affected by evolutionary forces within major lineages remains unclear. Here, we report a comprehensive comparison of 3D genomes, using high resolution Hi-C data in fibroblast cells of fish, chickens, and 10 mammalian species.ResultsThis analysis shows a correlation between genome size and chromosome length that affects chromosome territory (CT) organization in the upper hierarchy of genome architecture, whereas lower hierarchical features, including local transcriptional availability of DNA, are selected through the evolution of vertebrates. Furthermore, conservation of topologically associating domains (TADs) appears strongly associated with the modularity of expression profiles across species. Additionally, LINE and SINE transposable elements likely contribute to heterochromatin and euchromatin organization, respectively, during the evolution of genome architecture.ConclusionsOur analysis uncovers organizational features that appear to determine the conservation and transcriptional regulation of functional genes across species. These findings can guide ongoing investigations of genome evolution by extending our understanding of the mechanisms shaping genome architecture.Supplementary InformationThe online version contains supplementary material available at 10.1186/s12915-022-01301-7.

HiCARN: resolution enhancement of Hi-C data using cascading residual networks.

MotivationHigh throughput chromosome conformation capture (Hi-C) contact matrices are used to predict 3D chromatin structures in eukaryotic cells. High-resolution Hi-C data are less available than low-resolution Hi-C data due to sequencing costs but provide greater insight into the intricate details of 3D chromatin structures such as enhancer–promoter interactions and sub-domains. To provide a cost-effective solution to high-resolution Hi-C data collection, deep learning models are used to predict high-resolution Hi-C matrices from existing low-resolution matrices across multiple cell types.ResultsHere, we present two Cascading Residual Networks called HiCARN-1 and HiCARN-2, a convolutional neural network and a generative adversarial network, that use a novel framework of cascading connections throughout the network for Hi-C contact matrix prediction from low-resolution data. Shown by image evaluation and Hi-C reproducibility metrics, both HiCARN models, overall, outperform state-of-the-art Hi-C resolution enhancement algorithms in predictive accuracy for both human and mouse 1/16, 1/32, 1/64 and 1/100 downsampled high-resolution Hi-C data. Also, validation by extracting topologically associating domains, chromosome 3D structure and chromatin loop predictions from the enhanced data shows that HiCARN can proficiently reconstruct biologically significant regions.Availability and implementationHiCARN can be accessed and utilized as an open-sourced software at: https://github.com/OluwadareLab/HiCARN and is also available as a containerized application that can be run on any platform.Supplementary information Supplementary data are available at Bioinformatics online.

Comprehensive 3D epigenomic maps define limbal stem/progenitor cell function and identity

The insights into how genome topology couples with epigenetic states to govern the function and identity of the corneal epithelium are poorly understood. Here, we generate a high-resolution Hi-C interaction map of human limbal stem/progenitor cells (LSCs) and show that chromatin multi-hierarchical organisation is coupled to gene expression. By integrating Hi-C, epigenome and transcriptome data, we characterize the comprehensive 3D epigenomic landscapes of LSCs. We find that super-silencers mediate gene repression associated with corneal development, differentiation and disease via chromatin looping and/or proximity. Super-enhancer (SE) interaction analysis identified a set of SE interactive hubs that contribute to LSC-specific gene activation. These active and inactive element-anchored loop networks occur within the cohesin-occupied CTCF-CTCF loops. We further reveal a coordinated regulatory network of core transcription factors based on SE-promoter interactions. Our results provide detailed insights into the genome organization principle for epigenetic regulation of gene expression in stratified epithelia.