Biology subjects

Hughes, J. R.

Publications and source records attributed to Hughes, J. R..

36 records · Page 2Linked to original sources

bioRxiv2020

A functional overlap between actively transcribed genes and chromatin boundary elements

Mammalian genomes are subdivided into large (50-2000 kb) regions of chromatin referred to as Topologically Associating Domains (TADs or sub-TADs). Chromatin within an individual TAD contacts itself more frequently than with regions in surrounding TADs thereby directing enhancer-promoter interactions. In many cases, the borders of TADs are defined by convergently orientated boundary elements associated with CCCTC-binding factor (CTCF), which stabilises the cohesin complex on chromatin and prevents its translocation. This delimits chromatin loop extrusion which is thought to underlie the formation of TADs. However, not all CTCF-bound sites act as boundaries and, importantly, not all TADs are flanked by convergent CTCF sites. Here, we examined the CTCF binding sites within a [~]70 kb sub-TAD containing the duplicated mouse -like globin genes and their five enhancers (5-R1-R2-R3-Rm-R4-1-2-3). The 5 border of this sub-TAD is defined by a pair of CTCF sites. Surprisingly, we show that deletion of the CTCF binding sites within and downstream of the -globin locus leaves the sub-TAD largely intact. The predominant 3 border of the sub-TAD is defined by a steep reduction in contacts: this corresponds to the transcribed 2-globin gene rather than the CTCF sites at the 3-end of the sub-TAD. Of interest, the almost identical 1- and 2-globin genes interact differently with the enhancers, resulting in preferential expression of the proximal 1-globin gene which behaves as a partial boundary between the enhancers and the distal 2-globin gene. Together, these observations provide direct evidence that actively transcribed genes can behave as boundary elements. Significance StatementMammalian genomes are complex, organised 3D structures, partitioned into Topologically Associating Domains (TADs): chromatin regions that preferentially self-interact. These chromatin interactions are thought to be driven by a mechanism that continuously extrudes chromatin loops, forming structures delimited by chromatin boundary elements and reflecting the activity of enhancers and promoters. Boundary elements bind architectural proteins such as CCCTC-binding factor (CTCF). Previously, an overlap between the functional roles of enhancers and promoters has been shown. However, whether there is overlap between enhancers/promoters and boundary elements is not known. Here, we show that actively transcribed genes can also behave as boundary elements, similar to CTCF boundaries. In both cases, multi-protein complexes bound to these regions may stall the process of chromatin loop extrusion.

genomics↗

bioRxiv2020

Multi Locus View : An Extensible Web Based Tool for the Analysis of Genomic Data

MotivationTracking and understanding data quality, analysis and reproducibility are critical concerns in the biological sciences. This is especially true in genomics where Next Generation Sequencing (NGS) based technologies such as ChIP-seq, RNA-seq and ATAC-seq are generating a flood of genome-scale data. These data-types are extremely high level and complex with single experiments capable of mapping ten to hundreds of thousands of biologically meaningful events across the genome. However, such data are usually processed with automated tools and pipelines, generating tabular outputs and static visualizations. These are difficult to interact with and require substantial bioinformatic skills to manipulate and query. Similarly, interpretation is normally made at a high level without the ability to visualise the underlying data in detail and so the complexity and quality of the real underlying biological signal is lost. Also genomics datasets require integration with other genomics datasets to be properly interpreted and this integration with multiple tracks again requires substantial bioinformatics skills and is difficult to visualise across multiple pertinent datasets. Conventional genome browsers do allow for the detailed visualisation of multiple tracks but are limited to browsing single locations and do not allow for interactions with the dataset as a whole. MLV has been developed to allow users to fluidly interact with genomics datasets at multiple scales, from complete metadata labelled and clustered populations to detailed representations of individual elements. It has inbuilt tools to integrate signals across multiple datasets and to perform dimensionality reduction and clustering analysis based on the extracted signal, allowing for the high-level analysis of complex datasets while maintaining visualisation of the fine grain structure of the data. MLVs ability to visualise clustering within the data combined with efficient tools for large-scale tagging of individual elements makes it a unique tool for the generation of annotated datasets for modern machine learning approaches. ResultsMulti Locus View (MLV) is a web based tool for the visualisation, analysis and annotation of Next Generation Sequencing data sets. The user is able to browse the raw data, cluster, and combine the data with other analysis. Intuitive filtering and visualisation then enables the user to quickly locate and annotate regions of interest. User datasets can then be shared with other users or made public for quick assessment from the academic community. MLV is publically available at https://mlv.molbiol.ox.ac.uk and the source code is available at https://github.com/Hughes-Genome-Group/mlv

bioinformatics↗

bioRxiv2020

Targeted high-resolution chromosome conformation capture at genome-wide scale.

Chromosome conformation capture (3C) provides an adaptable tool for studying diverse biological questions. Current 3C methods provide either low-resolution interaction profiles across the entire genome, or high-resolution interaction profiles at up to several hundred loci. All 3C methods are affected to varying degrees by inefficiency, bias and noise. As such, generation of reproducible high-resolution interaction profiles has not been achieved at scale. To overcome this barrier, we systematically tested and improved upon current methods. We show that isolation of 3C libraries from intact nuclei, as well as shortening and titration of enrichment oligonucleotides used in high-resolution methods reduces noise and increases on-target sequencing. We combined these technical modifications into a new method Nuclear-Titrated (NuTi) Capture-C, which provides a >3-fold increase in informative sequencing content over current Capture-C protocols. Using NuTi Capture-C we target 8,061 promoters in triplicate, demonstrating that this method generates reproducible high-resolution genome-wide 3C interaction profiles at scale.

genomics↗

bioRxiv2020

CaptureCompendium: a comprehensive toolkit for 3C analysis.

DNA folding within nuclei is a highly ordered process, with implications for gene regulation and development. An array of chromosome conformation capture (3C) methods have been developed to investigate how DNA is packaged within nuclei and to interrogate specific interactions. While these methods use different approaches to examine target loci (many-versus-all) or the entire genome (all-versus-all), they all rely on the core principle of endonuclease digestion and proximity-based ligation to re-arrange genomic order to reflect the three-dimensional nuclear conformation. This sequence reorganization creates novel chimeric DNA fragments which require specialist bioinformatic tools to analyze and visualize. Despite this need for specialist bioinformatic skills, the core biological importance of genome folding has seen widespread methodological uptake. To service the needs of experimentalists using the many-versus-all Capture-C family of methods we have developed CaptureCompendium; a toolkit of software to simplify the design, analysis and presentation of 3C experiments.

bioinformatics↗

bioRxiv2019

Enhancers predominantly regulate gene expression in vivo via transcription initiation.

Gene transcription occurs via a cycle of linked events including initiation, promoter proximal pausing and elongation of RNA polymerase II (Pol II). A key question is how do transcriptional enhancers influence these events to control gene expression? Here we have used a new approach to quantify transcriptional initiation and pausing in vivo, while simultaneously identifying transcription start sites (TSSs) and pause-sites (TPSs) from single RNA molecules. When analyzed in parallel with nascent RNA-seq, these data show that differential gene expression is achieved predominantly via changes in transcription initiation rather than Pol II pausing. Using genetically engineered mouse models deleted for specific enhancers we show that these elements control gene expression via Pol II recruitment and/or initiation rather than via promoter proximal pause release. Together, our data show that enhancers, in general, control gene expression predominantly by Pol II recruitment and initiation rather than via pausing.

genomics↗

bioRxiv2019

An integrated platform to systematically identify causal variants and genes for polygenic human traits.

Genome-wide association studies (GWAS) have identified over 150,000 links between common genetic variants and human traits or complex diseases. Over 80% of these associations map to polymorphisms in non-coding DNA. Therefore, the challenge is to identify disease-causing variants, the genes they affect, and the cells in which these effects occur. We have developed a platform using ATAC-seq, DNaseI footprints, NG Capture-C and machine learning to address this challenge. Applying this approach to red blood cell traits identifies a significant proportion of known causative variants and their effector genes, which we show can be validated by direct in vivo modelling.

genetics↗

bioRxiv2019

Absolute quantification of transcription factors reveals principles of gene regulation in erythropoiesis

Dynamic cellular processes such as differentiation are driven by changes in the abundances of transcription factors (TFs). Yet, despite years of studies we still do not know the protein copy number of TFs in the nucleus. Here, by determining the absolute abundances of 103 TFs and co-factors during the course of human erythropoiesis, we provide a dynamic and quantitative scale for TFs in the nucleus. Furthermore, we establish the first Gene Regulatory Network of cell fate commitment that integrates temporal protein stoichiometry data with mRNA measurements. The model revealed quantitative imbalances in TFs cross-antagonistic relationships that underlie lineage determination. Finally, we made the surprising discovery that in the nucleus, corepressors are dramatically more abundant than coactivators at the protein, but not at the RNA level, with profound implications for understanding transcriptional regulation. These analyses provide a unique quantitative framework to understand transcriptional regulation of cell differentiation in a dynamic context.

systems biology↗

bioRxiv2019

Dissection of the 4D chromatin structure of the α-globin locus through in vivo erythroid differentiation with extreme spatial and temporal resolution

Mammalian gene expression patterns are controlled by regulatory elements, which interact within Topologically Associating Domains (TADs). The relationship between activation of regulatory elements, formation of structural chromatin interactions and gene expression during development is unclear. We developed Tiled-C, a low-input Chromosome Conformation Capture (3C) approach, to study chromatin architecture at high spatial and temporal resolution through in vivo mouse erythroid differentiation. Integrated analysis of matched chromatin accessibility and single-cell expression data shows that regulatory elements gradually become accessible within pre-existing TADs during early differentiation. This is followed by structural re-organization within the TAD and formation of specific contacts between enhancers and promoters. In contrast to previous reports, our high-resolution data show that these enhancer-promoter interactions are not established prior to gene expression, but formed gradually during differentiation, concomitant with progressive upregulation of gene activity. Together, these results provide new insight into the close, interdependent relationship between chromatin architecture and gene regulation during development.

genomics↗

bioRxiv2019

Modelling erythropoiesis in congenital dyserythropoietic anaemia type I (CDA-I)

We employ and extensively characterise an ex vivo culture system to study terminal erythroid maturation of CD34+ progenitors from the peripheral blood of normal individuals and patients with Congenital Dyserythropoietic Anaemia type 1 (CDA-I). Using morphological analysis, FACS analysis and the proteomic approach CyTOF, we analysed patient-derived erythroblasts stage-matched with those from healthy donors during the expansion phase and into early differentiation. In patient cells, aspects of disordered erythropoiesis manifest midway through differentiation, including increased proliferation and changes in the DNA accessibility profile. We also show that cultured erythroblasts from CDA-I patients recapitulate the pathognomic feature of this erythroid disorder with up to 40% of the cells having abnormal spongy chromatin morphology by electron microscopy, as well as upregulation of GDF15, a marker of ineffective erythropoiesis. In the tertiary phase of culture, patient cells show significantly less enucleation and there is persistence of earlier erythroid precursors. Furthermore, the enucleation defect appears to be more severe in patients with mutations in C15orf41, as compared to the other known causative gene CDAN1, indicating a genotype/phenotype correlation in CDA-I. Such erythroblasts are a valuable resource for investigating the pathogenesis of this disease and provide the opportunity for streamlining diagnosis for CDA-I patients and ultimately other forms of unexplained anaemia.

genetics↗

bioRxiv2019

DeepC: Predicting chromatin interactions using megabase scaled deep neural networks and transfer learning.

Understanding 3D genome structure requires high throughput, genome-wide approaches. However, assays for all vs. all chromatin interaction mapping are expensive and time consuming, which severely restricts their usage for large-scale mutagenesis screens or for mapping the impact of sequence variants. Computational models sophisticated enough to grasp the determinants of chromatin folding provide a unique window into the functional determinants of 3D genome structure as well as the effects of genome variation.\n\nA chromatin interaction predictor should work at the base pair level but also incorporate large-scale genomic context to simultaneously capture the large scale and intricate structures of chromatin architecture. Similarly, to be a flexible and generalisable approach it should also be applicable to data it has not been explicitly trained on. To develop a model with these properties, we designed a deep neuronal network (deepC) that utilizes transfer learning to accurately predict chromatin interactions from DNA sequence at megabase scale. The model generalizes well to unseen chromosomes and works across cell types, Hi-C data resolutions and a range of sequencing depths. DeepC integrates DNA sequence context on an unprecedented scale, bridging the different levels of resolution from base pairs to TADs. We demonstrate how this model allows us to investigate sequence determinants of chromatin folding at genome-wide scale and to predict the importance of regulatory elements and the impact of sequence variations.

genomics↗

bioRxiv2019

A revised model for promoter competition based on multi-way chromatin interactions

Specific communication between gene promoters and enhancers is critical for accurate regulation of gene expression. However, it remains unclear how specific interactions between multiple regulatory elements and genes contained within a single chromatin domain are coordinated. Recent technological advances allow for the investigation of multi-way chromatin interactions at single alleles in individual nuclei. This can provide insights into how multiple regulatory elements cooperate or compete for transcriptional activation. We have used these techniques in a mouse model in which the -globin domain is extended to include several additional genes. This allows us to determine how the interactions of the -globin super-enhancer are distributed between multiple promoters in a single domain. Our data show that gene promoters do not form mutually exclusive interactions with the super-enhancer, but all interact simultaneously in a single complex. These finding show that promoters within the same domain do not structurally compete for interactions with enhancers, but form a regulatory hub structure, consistent with the recent model of transcriptional activation in phase-separated nuclear condensates.

genomics↗

bioRxiv2019

CSynth: A Dynamic Modelling and Visualisation Tool for 3D Chromatin Structure

The 3D structure of chromatin in the nucleus is important for gene expression and regulation. Chromosomal conformation capture techniques, such as Hi-C, generate large amounts of data showing interaction points on the genome but these are hard to interpret using standard tools. We have developed CSynth, a high performance 3D genome browser and real time chromatin restraint-based modeller to visualise dynamic and interactive models of chromatin capture data. CSynth does its calculations in the GPU hence is much faster than existing modelling software to infer and visualise the chromatin structure which also allow real-time interaction with the modelling parameters. It also allows straightforward comparison of interaction data and the results of third party 3D modelling outputs. In addition we include an option to view and manipulate these complicated structures using Virtual Reality (VR) allowing scientists to immerse themselves in the models for further understanding. This VR component has also proven to be a valuable teaching and public engagement tool. CSynth is web based and available to use at http://csynth.org.

bioinformatics↗

bioRxiv2018

NGseqBasic - a single-command UNIX tool for ATAC-seq, DNaseI-seq, Cut-and-Run, and ChIP-seq data mapping, high-resolution visualisation, and quality control

With decreasing cost of next-generation sequencing (NGS), we are observing a rapid rise in the volume of big data in academic research, healthcare and drug discovery sectors. The present bottleneck for extracting value from these big data sets is data processing and analysis. Considering this, there is still a lack of reliable, automated and easy to use tools that will allow experimentalists to assess the quality of the sequenced libraries and explore the data first hand, without the need of investing a lot of time of computational core analysts in the early stages of analysis.\n\nNGseqBasic is an easy-to-use single-command analysis tool for chromatin accessibility (ATAC, DNaseI) and ChIP sequencing data, providing support to also new techniques such as low cell number sequencing and Cut-and-Run. It takes in fastq, fastq.gz or bam files, conducts all quality control, trimming and mapping steps, along with quality control and data processing statistics, and combines all this to a single-click loadable UCSC data hub, with integral statistics html page providing detailed reports from the analysis tools and quality control metrics. The tool is easy to set up, and no installation is needed. A wide variety of parameters are provided to fine-tune the analysis, with optional setting to generate DNase footprint or high resolution ChIP-seq tracks. A tester script is provided to help in the setup, along with a test data set and downloadable example user cases.\n\nNGseqBasic has been used in the routine analysis of next generation sequencing (NGS) data in high-impact publications 1,2. The code is actively developed, and accompanied with Git version control and Github code repository. Here we demonstrate NGseqBasic analysis and features using DNaseI-seq data from GSM689849, and CTCF-ChIP-seq data from GSM2579421, as well as a Cut-and-Run CTCF data set GSM2433142, and provide the one-click loadable UCSC data hubs generated by the tool, allowing for the ready exploration of the run results and quality control files generated by the tool.\n\nAvailabilityDownload, setup and help instructions are available on the NGseqBasic web site http://userweb.molbiol.ox.ac.uk/public/telenius/NGseqBasicManual/external/\n\nBioconda users can load the tool as library \"ngseqbasic\". The source code with Git version control is available in https://github.com/Hughes-Genome-Group/NGseqBasic/releases.\n\nContactjelena.telenius@imm.ox.ac.uk

bioinformatics↗

bioRxiv2018

DOT1L inhibition reveals a distinct class ofenhancers dependent on H3K79 methylation

Enhancer elements are a key regulatory feature of many important genes. Several general features including the presence of specific histone modifications are used to identify and subcategorize enhancers. Here we identify a distinct subset of enhancers in leukemia cells that are functionally dependent upon H3K79me3. Using the DOT1L inhibitor, EPZ-5676, we show that loss of H3K79me3 at these H3K79me3 enhancer elements (KEEs) leads to reduced chromatin accessibility, histone acetylation and transcription factor binding. We then use Capture-C, a high-resolution chromosome conformation capture technique, to show that H3K79me3 is required for KEE interactions with the promoter as well as transcription of the associated genes. Together these data implicate H3K79me3 in having a functional role at a subset of active enhancers where it helps maintain histone acetylation and chromatin accessibility, potentially by promoting phase-separated condensates.

molecular biology↗

bioRxiv2018

Single-cell chromatin interactions reveal regulatory hubs in dynamic compartmentalized domains

The promoters of mammalian genes are commonly regulated by multiple distal enhancers, which physically interact within discrete chromatin domains. How such domains form and how the regulatory elements within them interact within single cells is not understood. To address this we developed Tri-C, a new Chromosome Conformation Capture (3C) approach to identify concurrent chromatin interactions at individual alleles within single cells. The heterogeneity of interactions observed between such cells shows that CTCF-mediated formation of chromatin domains and interactions within them are dynamic processes. Importantly, our analyses reveal higher-order structures involving simultaneous interactions between multiple enhancers and promoters within individual cells. This provides a structural basis for understanding how multiple cis-elements act together to establish robust regulation of gene expression.

genetics↗

bioRxiv2017

A tissue-specific self-interacting chromatin domain forms independently of enhancer-promoter interactions

A variety of self-interacting domains, defined at different levels of resolution, have been described in mammalian genomes. These include Chromatin Compartments (A and B)1, Topologically Associated Domains (TADs)2,3, contact domains4,5, sub-TADs6, insulated neighbourhoods7 and frequently interacting regions (FIREs)8. Whereas many studies have found the organisation of self-interacting domains to be conserved across cell types389, some do form in a lineage-specific manner6710. However, it is not clear to what degree such tissue-specific structures result from processes related to gene activity such as enhancer-promoter interactions or whether they form earlier during lineage commitment and are therefore likely to be prerequisite for promoting gene expression. To examine these models of genome organisation in detail, we used a combination of high-resolution chromosome conformation capture, a newly-developed form of quantitative fluorescence in-situ hybridisation and super-resolution imaging to study a 70 kb self-interacting domain containing the mouse -globin locus. To understand how this self-interacting domain is established, we studied the region when the genes are inactive and during erythroid differentiation when the genes are progressively switched on. In contrast to many current models of long-range gene regulation, we show that an erythroid-specific, decompacted self-interacting domain, delimited by convergent CTCF/cohesin binding sites, forms prior to the onset of robust gene expression. Using previously established mouse models we show that formation of the self-interacting domain does not rely on interactions between the -globin genes and their enhancers. As there are also no tissue-specific changes in CTCF binding, then formation of the domain may simply depend on the presence of activated lineage-specific cis-elements driving a transcription-independent mechanism for opening chromatin throughout the 70 kb region to create a permissive environment for gene expression. These findings are consistent with a model of loop-extrusion in which all segments of chromatin, within a region delimited by CTCF boundary elements, can contact each other. Our findings suggest that activation of tissue-specific element(s)within such a self-interacting region is sufficient to influence all chromatin within the domain.

cell biology↗

bioRxiv2017

Development of a Self-Report Measure of Reward Sensitivity: A Test in Current and Former Smokers

IntroductionTobacco use or abstinence may increase or decrease reward sensitivity. Most existing measures of reward sensitivity were developed decades ago, and few have undergone extensive psychometric testing.\n\nMethodsWe developed a 58-item survey of the anticipated enjoyment from, wanting for, and frequency of common rewards (the Rewarding Events Inventory - REI). The current analyses focuses on ratings of anticipated enjoyment. The first validation study recruited current and former smokers from internet sites. The second study recruited smokers who wished to quit and monetarily reinforced them to stay abstinent in a laboratory study, and a comparison group of former smokers. In both studies, participants completed the inventory on two occasions, 3-7 days apart. They also completed four anhedonia scales and a behavioral test of reduced reward sensitivity.\n\nResultsHalf of the enjoyment ratings loaded on four factors: socializing, active hobbies, passive hobbies, and sex/drug use. Cronbach alpha coefficients were all [≥] 0.73 for overall mean and factor scores. Test-retest correlations were all [≥] 0.83. Correlations of the overall and factor scores with frequency of rewards, anhedonia scales were 0.19 - 0.53, except for the sex/drugs factor. The scores did not correlate with behavioral tests of reward and did not differ between current and former smokers. Lower overall mean enjoyment score predicted a shorter time to relapse.\n\nDiscussionInternal reliability and test-retest reliability of the enjoyment outcomes of the REI are excellent, and construct and predictive validity are modest but promising. The REI is comprehensive and up-to-date, yet is short enough to use on repeated occasions. Replication tests, especially predictive validity tests, are needed.\n\nImplicationsBoth use of and abstinence from nicotine appears to increase or decrease how rewarding non-drug rewards are; however, self-report scales to test this have limitations. Our inventory of enjoyment from 58 rewards appears to be reliable and valid as well as comprehensive and up-to-date, yet is short enough to use on repeated occasions. Replication tests, especially of the predictive validity of our scale, are needed.

neuroscience↗

bioRxiv2017

Does tobacco abstinence decrease reward sensitivity? A human laboratory test

IntroductionAnimal studies report abstinence from nicotine makes rewards less rewarding; however, the results of human tests of the effects of cessation on reward sensitivity are mixed. The current study tested reward sensitivity in abstinent smokers using more rigorous methods than most prior studies.\n\nMethodsA human laboratory study compared outcomes for 1 week prior to quitting to those during 4 weeks post-quit. The study used smokers trying to quit, objective and subjective measures, multiple measures during smoking and abstinence, and monetary rewards to increase the prevalence of abstinence. Current daily smokers (n = 211) who were trying to quit completed an operant measure of reward sensitivity and a survey of pleasure from various rewards as well as self-reports of anhedonia, delay discounting, positive affect and tobacco withdrawal twice each week. A comparison group of long-term former smokers (n = 67) also completed the tasks weekly for 4 weeks. Primary analyses were based on the 61 current smokers who abstained for all 4 weeks.\n\nResultsStopping smoking decreased self-reported pleasure from rewards but did not decrease reward sensitivity on the operant task. Abstinence also decreased self-reported reward frequency and increased the two anhedonia measures. However, the changes with abstinence were small for all outcomes (6-14%) and most lasted less than a week.\n\nConclusionAbstinence from tobacco decreased most self-report measures of reward sensitivity; however, it did not change the objective measure. The self-report effects were small.\n\nImplicationsO_LIAnimal research suggests that nicotine withdrawal decreases reward sensitivity. Replication tests of this in humans have produced inconsistent results.\nC_LIO_LIWe report what we believe is a more rigorous test\nC_LIO_LIWe found smoking abstinence slightly decreases self-reports of reward sensitivity but does not do so for behavioral measures of reward sensitivity\nC_LI

neuroscience↗