The Past and The Present of the 4D Nucleome Project

The 4D Nucleome project (Dekker et al., 2017) aims to understand the principles underlying nuclear organization (the 3D folding of the human genome); that is, the researchers behind this colossal undertaking hope to understand the underlying nuclear organization in space and time, the role of nuclear organization in gene expression and cell function, and the changes in nuclear organization that underpin normal and pathological development. A recent paper from the 4D Nucleome Project – which integrates the herculean efforts made from researchers from the labs of Job Dekker, Xiaotao Wang, Mario Nicodemi, Bing Ren, Sheng Zhong, Jennifer E. Phillips-Cremins, David M. Gilbert, Katherine S. Pollard, Frank Alber, Jian Ma, William S. Noble, and Feng Yue – provides an update on their work, which describes the 4D nucleome in two distinct cell types - human embryonic stem cells and immortalized fibroblasts (Dekker et al.).

Overall, the integration of diverse genomic data describing the 4D nucleome enabled the generation of detailed classifications/annotations of chromosomal domain types and their subnuclear positions and single-cell models of the 3D nuclear environment for all genes. Furthermore, benchmarking helped to describe the strengths of the distinct assays employed (and provided guidelines for future studies), while models of population-based and individual cell-to-cell variation in genome structure highlighted connections between chromosome folding, nuclear organization, chromatin looping, gene transcription, and DNA replication

Paired-Tag technology from Epigenome Technologies generates joint epigenetic and transcriptomic profiles at single-cell resolution and detects histone modifications and RNA transcripts in individual nuclei with efficiency comparable to single-nucleus RNA-seq/ChIP-seq assays. Could the integration of datasets generated by this technology provide another layer of epigenetic information, furthering our understanding of the 3D folding of the human genome?

The Ongoing Second Phase of the 4D Nucleome Project: A Brief Overview

While a simple blog post cannot unpack the wealth of data presented in this paper, we bring you a brief rundown of the most salient points from this massive undertaking! The study first compared distinct genomic assays to determine and quantify the features of the 4D nucleome. A comparative analysis of assays used to generate contact maps revealed that a variety of methods captured similar patterns of chromatin organization; however, the team identified the chromosome conformation capture-based assays with the largest and smallest dynamic ranges. A similar comparison highlighted the overall utility of each method studied for qualitatively detecting compartmentalization and identifying compartment domains; however, large quantitative differences remain regarding compartment domain size, the detection of smaller compartment domains, and quantifying the strength of compartmentalization. Analysis of the assays that detect topologically associating domain boundaries revealed that a variety of chromatin interaction assays robustly detected local domain boundary formation as a feature of genome folding. Finally, the comparative analyses of chromatin loop detection methods demonstrated that distinct assays preferentially detected distinct types of chromatin loops; while certain assays provided an unbiased view of genome architecture and effectively capture CTCF/cohesin-mediated insulator loops, targeted methods displayed enrichment for transcription-related loops and sensitivity to poised promoter loops.

A previous publication had demonstrated that integrating complementary 3D genome-mapping data into a unified probabilistic model – SPIN (Wang et al.) - permitted the derivation of linear genome-wide annotations of spatial nuclear compartments. Annotations – denoted as SPIN states - revealed distinct spatial localization patterns of loci relative to nuclear bodies and displayed robust connections between large-scale chromosome structure and function. The authors of this new study aimed to identify primary SPIN states in the two cell types involved; their initial analyses demonstrated that, while SPIN states correlated with transcriptionally permissive histone modifications, they also associated with a cell-type-specific distribution of repressive histone modifications. A subsequent comparison of SPIN states with distinct types of chromatin-associated RNAs revealed that the genomic target sequences of diverse types of repeat sequences associated with distinct types displayed enrichment for distinct SPIN states. Overall, these results demonstrated that the SPIN framework integrated nuclear organization mapping data to produce genome-wide large-scale compartmentalization patterns relative to multiple nuclear bodies, with SPIN states stratifying orthogonal functional genomic data.

The next phase of the paper employed multiple datasets and an integrative genome modelling platform (Su et al.) to generate a 3D view of the genome in the two cell types under study and then characterize the nuclear microenvironment of loci. Analysis of the nuclear microenvironment of chromatin across SPIN states revealed distinct enrichments of 3D structure features. Furthermore, comparing genome structure across cell types revealed that genes with large expression differences often localize to distinct nuclear microenvironments, whereas genes highly expressed in both cell types typically localize to similar microenvironments. Meanwhile, analysis of gene expression highlighted a general correlation with the nuclear microenvironment (with some exceptions).

The subsequent analysis of variability in single-cell 3D genome structure, using different integrative modelling approaches, revealed that contact maps from single-cell and bulk chromosome conformation capture/next-generation sequencing data yielded comparable results. While these findings highlighted cell-to-cell variation in chromatin folding within individual cells, they also underscored the presence of loop formation, topologically associating domain-like domains, and compartments at the single-cell and single-molecule levels, suggesting that these chromatin folding activities do not merely reflect correlated properties of chromatin folding observed in bulk contact maps. Using this data, the authors examined the relationship between the number of distal enhancers linked to promoters and the transcription levels of corresponding protein-coding genes, explored the relationship between housekeeping gene expression and enhancers, evaluated the formation of enhancer–promoter loops near the lamina, and evaluated functional domains at different scales. This final analytical step revealed that the relationship between higher-order chromatin structure and function depended on the length scale of the folding feature.

4D Nucleome Project So Far: A Summary

This second phase of the 4D Nucleome Project focused on integrating genomic with imaging data, the development and application of multi-omic single-cell datasets, and the analysis of alterations to the 4D nucleome; as such, they provide a detailed view of the human 4D nucleome thanks to the integration of data obtained with a range of genomic methods. These data reveal how each method quantitatively contributed to delineating unique and common aspects of genome folding and defining connections between chromosome folding and looping, nuclear positioning, proximity to nuclear bodies, cell-to-cell variation in organization, and genomic functions. These data provide a detailed view of the human genome organized within cells and lay the foundation for future deep exploration of the structure and function of genomes in normal and disease states. Encouragingly, the teams behind this study have made the described datasets publicly available at the 4D Nucleome Data Coordination and Integration Center (https://data.4dnucleome.org/).

The implementation of Paired-Tag technology from Epigenome Technologies, which generates joint epigenetic and transcriptomic profiles at single-cell resolution and detects histone modifications and RNA transcripts in individual nuclei with efficiency comparable to single-nucleus RNA-seq/ChIP-seq assays, has the potential to provide deeper insight into such research aims. What more could the simultaneous single-cell analysis of histone modification and transcriptomic profiles tell us about the 3D folding of the human genome?