Detecting chromosomal interactions in Capture Hi-C data with CHiCAGO and companion tools

Using paired-end read orientations to assess technical biases in capture Hi-C

Hi-C and capture Hi-C (CHi-C) both leverage paired-end sequencing of chimeric fragments to gauge the strength of interactions based on the total number of paired-end reads mapped to a common pair of restriction fragments. Mapped paired-end reads can have four relative orientations, depending on the genomic positions and strands of the two reads. We assigned one paired-end read orientation to each of the four possible re-ligations that can occur between two given restriction fragments. In a large hematopoietic cell dataset, we determined the read pair counts of interactions separately for each orientation. Interactions with imbalances in the counts occur much more often than expected by chance for both Hi-C and CHi-C. Based on such imbalances, we identified target restriction fragments enriched at only one instead of both ends. By matching them to the baits used for the experiments, we confirmed our assignment of paired-end read orientations and gained insights that can inform bait design. An analysis of unbaited fragments shows that, beyond bait effects, other known types of technical biases are reflected in count imbalances. Taking advantage of distance-dependent contact frequencies, we assessed the impact of such biases. Our results have the potential to improve the design and interpretation of CHi-C experiments.

Resolving the three-dimensional interactome of Human Accelerated Regions during human and chimpanzee neurodevelopment

Human Accelerated Regions (HARs) are highly conserved across species but exhibit a significant excess of human-specific sequence changes, suggesting they may have gained novel functions in human evolution. HARs include transcriptional enhancers with human-specific activity and have been implicated in the evolution of the human brain. However, our understanding of how HARs contributed to uniquely human features of the brain is hindered by a lack of insight into the genes and pathways that HARs regulate. It is unclear whether HARs acted by altering the expression of gene targets conserved between HARs and their chimpanzee orthologs or by gaining new gene targets in human, a mechanism termed enhancer hijacking. We generated a high-resolution map of chromatin interactions for 1,590 HARs and their orthologs in human and chimpanzee neural stem cells (NSCs) to comprehensively identify gene targets in both species. HARs and their chimpanzee orthologs targeted a conserved set of 2,963 genes enriched for neurodevelopmental processes including neurogenesis and synaptic transmission. Changes in HAR enhancer activity were correlated with changes in conserved gene target expression. Conserved targets were enriched among genes differentially expressed between human and chimpanzee NSCs or between human and non-human primate developing and adult brain. Species-specific HAR gene targets did not converge on known biological functions and were not significantly enriched among differentially expressed genes, suggesting that HARs did not alter gene expression via enhancer hijacking. HAR gene targets, including differentially expressed targets, also showed cell type-specific expression patterns in the developing human brain, including outer radial glia, which are hypothesized to contribute to human cortical expansion. Our findings support that HARs influenced human brain evolution by altering the expression of conserved gene targets and provide the means to functionally link HARs with novel human brain features.

Increased enhancer-promoter interactions during developmental enhancer activation in mammals

Remote enhancers are thought to interact with their target promoters via physical proximity, yet the importance of this proximity for enhancer function remains unclear. Here we investigate the three-dimensional (3D) conformation of enhancers during mammalian development by generating high-resolution tissue-resolved contact maps for nearly a thousand enhancers with characterized in vivo activities in ten murine embryonic tissues. Sixty-one percent of developmental enhancers bypass their neighboring genes, which are often marked by promoter CpG methylation. The majority of enhancers display tissue-specific 3D conformations, and both enhancer-promoter and enhancer-enhancer interactions are moderately but consistently increased upon enhancer activation in vivo. Less than 14% of enhancer-promoter interactions form stably across tissues; however, these invariant interactions form in the absence of the enhancer and are likely mediated by adjacent CTCF binding. Our results highlight the general importance of enhancer-promoter physical proximity for developmental gene activation in mammals.

Epigenomic mapping reveals distinct B cell acute lymphoblastic leukemia chromatin architectures and regulators

B cell lineage acute lymphoblastic leukemia (B-ALL) is composed of diverse molecular subtypes, and while transcriptional and DNA methylation profiling has been extensively examined, the chromatin landscape is not well characterized for many subtypes. We therefore mapped chromatin accessibility using ATAC-seq in primary B-ALL cells from 156 patients spanning ten molecular subtypes and present this dataset as a resource. Differential chromatin accessibility and transcription factor (TF) footprint profiling were employed and identified B-ALL cell of origin, TF-target gene interactions enriched in B-ALL, and key TFs associated with accessible chromatin sites preferentially active in B-ALL. We further identified over 20% of accessible chromatin sites exhibiting strong subtype enrichment and candidate TFs that maintain subtype-specific chromatin architectures. Over 9,000 genetic variants were uncovered, contributing to variability in chromatin accessibility among patient samples. Our data suggest that distinct chromatin architectures are driven by diverse TFs and inherited genetic variants that promote unique gene-regulatory networks.

MacroH2A restricts inflammatory gene expression in melanoma cancer-associated fibroblasts by coordinating chromatin looping

MacroH2A has established tumour suppressive functions in melanoma and other cancers, but an unappreciated role in the tumour microenvironment. Using an autochthonous, immunocompetent mouse model of melanoma, we demonstrate that mice devoid of macroH2A variants exhibit increased tumour burden compared with wild-type counterparts. MacroH2A-deficient tumours accumulate immunosuppressive monocytes and are depleted of functional cytotoxic T cells, characteristics consistent with a compromised anti-tumour response. Single cell and spatial transcriptomics identify increased dedifferentiation along the neural crest lineage of the tumour compartment and increased frequency and activation of cancer-associated fibroblasts following macroH2A loss. Mechanistically, macroH2A-deficient cancer-associated fibroblasts display increased myeloid chemoattractant activity as a consequence of hyperinducible expression of inflammatory genes, which is enforced by increased chromatin looping of their promoters to enhancers that gain H3K27ac. In summary, we reveal a tumour suppressive role for macroH2A variants through the regulation of chromatin architecture in the tumour stroma with potential implications for human melanoma.

A multiple super-enhancer region establishes inter-TAD interactions and controls Hoxa function in cranial neural crest

Enhancer-promoter interactions preferentially occur within boundary-insulated topologically associating domains (TADs), limiting inter-TAD interactions. Enhancer clusters in linear proximity, termed super-enhancers (SEs), ensure high target gene expression levels. Little is known about SE topological regulatory impact during craniofacial development. Here, we identify 2232 genome-wide putative SEs in mouse cranial neural crest cells (CNCCs), 147 of which target genes establishing CNCC positional identity during face formation. In second pharyngeal arch (PA2) CNCCs, a multiple SE-containing region, partitioned into Hoxa Inter-TAD Regulatory Element 1 and 2 (HIRE1 and HIRE2), establishes long-range inter-TAD interactions selectively with Hoxa2, that is required for external and middle ear structures. HIRE2 deletion in a Hoxa2 haploinsufficient background results in microtia. HIRE1 deletion phenocopies the full homeotic Hoxa2 knockout phenotype and induces PA3 and PA4 CNCC abnormalities correlating with Hoxa2 and Hoxa3 transcriptional downregulation. Thus, SEs can overcome TAD insulation and regulate anterior Hoxa gene collinear expression in a CNCC subpopulation-specific manner during craniofacial development.

3D genome mapping identifies subgroup-specific chromosome conformations and tumor-dependency genes in ependymoma

Ependymoma is a tumor of the brain or spinal cord. The two most common and aggressive molecular groups of ependymoma are the supratentorial ZFTA-fusion associated and the posterior fossa ependymoma group A. In both groups, tumors occur mainly in young children and frequently recur after treatment. Although molecular mechanisms underlying these diseases have recently been uncovered, they remain difficult to target and innovative therapeutic approaches are urgently needed. Here, we use genome-wide chromosome conformation capture (Hi-C), complemented with CTCF and H3K27ac ChIP-seq, as well as gene expression and DNA methylation analysis in primary and relapsed ependymoma tumors, to identify chromosomal conformations and regulatory mechanisms associated with aberrant gene expression. In particular, we observe the formation of new topologically associating domains ('neo-TADs') caused by structural variants, group-specific 3D chromatin loops, and the replacement of CTCF insulators by DNA hyper-methylation. Through inhibition experiments, we validate that genes implicated by these 3D genome conformations are essential for the survival of patient-derived ependymoma models in a group-specific manner. Thus, this study extends our ability to reveal tumor-dependency genes by 3D genome conformations even in tumors that lack targetable genetic alterations.

Genetic variation in cis-regulatory domains suggests cell type-specific regulatory mechanisms in immunity

Studying the interplay between genetic variation, epigenetic changes, and regulation of gene expression is crucial to understand the modification of cellular states in various conditions, including immune diseases. In this study, we characterize the cell-specificity in three key cells of the human immune system by building cis maps of regulatory regions with coordinated activity (CRDs) from ChIP-seq peaks and methylation data. We find that only 33% of CRD-gene associations are shared between cell types, revealing how similarly located regulatory regions provide cell-specific modulation of gene activity. We emphasize important biological mechanisms, as most of our associations are enriched in cell-specific transcription factor binding sites, blood-traits, and immune disease-associated loci. Notably, we show that CRD-QTLs aid in interpreting GWAS findings and help prioritize variants for testing functional hypotheses within human complex diseases. Additionally, we map trans CRD regulatory associations, and among 207 trans-eQTLs discovered, 46 overlap with the QTLGen Consortium meta-analysis in whole blood, showing that mapping functional regulatory units using population genomics allows discovering important mechanisms in the regulation of gene expression in immune cells. Finally, we constitute a comprehensive resource describing multi-omics changes to gain a greater understanding of cell-type specific regulatory mechanisms of immunity.

Low input capture Hi-C (liCHi-C) identifies promoter-enhancer interactions at high-resolution

Long-range interactions between regulatory elements and promoters are key in gene transcriptional control; however, their study requires large amounts of starting material, which is not compatible with clinical scenarios nor the study of rare cell populations. Here we introduce low input capture Hi-C (liCHi-C) as a cost-effective, flexible method to map and robustly compare promoter interactomes at high resolution. As proof of its broad applicability, we implement liCHi-C to study normal and malignant human hematopoietic hierarchy in clinical samples. We demonstrate that the dynamic promoter architecture identifies developmental trajectories and orchestrates transcriptional transitions during cell-state commitment. Moreover, liCHi-C enables the identification of disease-relevant cell types, genes and pathways potentially deregulated by non-coding alterations at distal regulatory elements. Finally, we show that liCHi-C can be harnessed to uncover genome-wide structural variants, resolve their breakpoints and infer their pathogenic effects. Collectively, our optimized liCHi-C method expands the study of 3D chromatin organization to unique, low-abundance cell populations, and offers an opportunity to uncover factors and regulatory networks involved in disease pathogenesis.

Hooked Up from a Distance: Charting Genome-Wide Long-Range Interaction Maps in Neural Cells Chromatin to Identify Novel Candidate Genes for Neurodevelopmental Disorders

DNA sequence variants (single nucleotide polymorphisms or variants, SNPs/SNVs; copy number variants, CNVs) associated to neurodevelopmental disorders (NDD) and traits often map on putative transcriptional regulatory elements, including, in particular, enhancers. However, the genes controlled by these enhancers remain poorly defined. Traditionally, the activity of a given enhancer, and the effect of its possible alteration associated to the sequence variants, has been thought to influence the nearest gene promoter. However, the obtainment of genome-wide long-range interaction maps in neural cells chromatin challenged this view, showing that a given enhancer is very frequently not connected to the nearest promoter, but to a more distant one, skipping genes in between. In this Perspective, we review some recent papers, who generated long-range interaction maps (by HiC, RNApolII ChIA-PET, Capture-HiC, or PLACseq), and overlapped the identified long-range interacting DNA segments with DNA sequence variants associated to NDD (such as schizophrenia, bipolar disorder and autism) and traits (intelligence). This strategy allowed to attribute the function of enhancers, hosting the NDD-related sequence variants, to a connected gene promoter lying far away on the linear chromosome map. Some of these enhancer-connected genes had indeed been already identified as contributive to the diseases, by the identification of mutations within the gene's protein-coding regions (exons), validating the approach. Significantly, however, the connected genes also include many genes that were not previously found mutated in their exons, pointing to novel candidate contributors to NDD and traits. Thus, long-range interaction maps, in combination with DNA variants detected in association with NDD, can be used as "pointers" to identify novel candidate disease-relevant genes. Functional manipulation of the long-range interaction network involving enhancers and promoters by CRISPR-Cas9-based approaches is beginning to probe for the functional significance of the identified interactions, and the enhancers and the genes involved, improving our understanding of neural development and its pathology.

Algorithmic considerations when analysing capture Hi-C data

Chromosome conformation capture methodologies have provided insight into the effect of 3D genomic architecture on gene regulation. Capture Hi-C (CHi-C) is a recent extension of Hi-C that improves the effective resolution of chromatin interactions by enriching for defined regions of biological relevance. The varying targeting efficiency between capture regions, however, introduces bias not present in conventional Hi-C, making analysis more complicated. Here we consider salient features of an algorithm that should be considered in evaluating the performance of a program used to analyse CHi-C data in order to infer meaningful interactions. We use the program CHICAGO to analyse promoter capture Hi-C data generated on 28 different cell lines as a case study.

3D genome organization links non-coding disease-associated variants to genes

Genome sequencing has revealed over 300 million genetic variations in human populations. Over 90% of variants are single nucleotide polymorphisms (SNPs), the remainder include short deletions or insertions, and small numbers of structural variants. Hundreds of thousands of these variants have been associated with specific phenotypic traits and diseases through genome wide association studies which link significant differences in variant frequencies with specific phenotypes among large groups of individuals. Only 5% of disease-associated SNPs are located in gene coding sequences, with the potential to disrupt gene expression or alter of the function of encoded proteins. The remaining 95% of disease-associated SNPs are located in non-coding DNA sequences which make up 98% of the genome. The role of non-coding, disease-associated SNPs, many of which are located at considerable distances from any gene, was at first a mystery until the discovery that gene promoters regularly interact with distal regulatory elements to control gene expression. Disease-associated SNPs are enriched at the millions of gene regulatory elements that are dispersed throughout the non-coding sequences of the genome, suggesting they function as gene regulation variants. Assigning specific regulatory elements to the genes they control is not straightforward since they can be millions of base pairs apart. In this review we describe how understanding 3D genome organization can identify specific interactions between gene promoters and distal regulatory elements and how 3D genomics can link disease-associated SNPs to their target genes. Understanding which gene or genes contribute to a specific disease is the first step in designing rational therapeutic interventions.

Capture-C: a modular and flexible approach for high-resolution chromosome conformation capture

Chromosome conformation capture (3C) methods measure the spatial proximity between DNA elements in the cell nucleus. Many methods have been developed to sample 3C material, including the Capture-C family of protocols. Capture-C methods use oligonucleotides to enrich for interactions of interest from sequencing-ready 3C libraries. This approach is modular and has been adapted and optimized to work for sampling of disperse DNA elements (NuTi Capture-C), including from low cell inputs (LI Capture-C), as well as to generate Hi-C like maps for specific regions of interest (Tiled-C) and to interrogate multiway interactions (Tri-C). We present the design, experimental protocol and analysis pipeline for NuTi Capture-C in addition to the variations for generation of LI Capture-C, Tiled-C and Tri-C data. The entire procedure can be performed in 3 weeks and requires standard molecular biology skills and equipment, access to a next-generation sequencing platform, and basic bioinformatic skills. Implemented with other sequencing technologies, these methods can be used to identify regulatory interactions and to compare the structural organization of the genome in different cell types and genetic models.

Prioritisation of Candidate Genes Underpinning COVID-19 Host Genetic Traits Based on High-Resolution 3D Chromosomal Topology

Genetic variants showing associations with specific biological traits and diseases detected by genome-wide association studies (GWAS) commonly map to non-coding DNA regulatory regions. Many of these regions are located considerable distances away from the genes they regulate and come into their proximity through 3D chromosomal interactions. We previously developed COGS, a statistical pipeline for linking GWAS variants with their putative target genes based on 3D chromosomal interaction data arising from high-resolution assays such as Promoter Capture Hi-C (PCHi-C). Here, we applied COGS to COVID-19 Host Genetic Consortium (HGI) GWAS meta-analysis data on COVID-19 susceptibility and severity using our previously generated PCHi-C results in 17 human primary cell types and SARS-CoV-2-infected lung carcinoma cells. We prioritise 251 genes putatively associated with these traits, including 16 out of 47 genes highlighted by the GWAS meta-analysis authors. The prioritised genes are expressed in a broad array of tissues, including, but not limited to, blood and brain cells, and are enriched for genes involved in the inflammatory response to viral infection. Our prioritised genes and pathways, in conjunction with results from other prioritisation approaches and targeted validation experiments, will aid in the understanding of COVID-19 pathology, paving the way for novel treatments.

Algorithmic considerations when analysing capture Hi-C data

Chromosome conformation capture methodologies have provided insight into the effect of 3D genomic architecture on gene regulation. Capture Hi-C (CHi-C) is a recent extension of Hi-C that improves the effective resolution of chromatin interactions by enriching for defined regions of biological relevance. The varying targeting efficiency between capture regions, however, introduces bias not present in conventional Hi-C, making analysis more complicated. Here we consider salient features of an algorithm that should be considered in evaluating the performance of a program used to analyse CHi-C data in order to infer meaningful interactions. We use the program CHICAGO to analyse promotor capture Hi-C data generated on 28 different cell lines as a case study.