Getting the most out of 3D genomics studies

The compaction of our genome in three-dimensional space is far from random, actually having a tightly controlled structure in space. This 3D structure is very important, because it provides a critical layer of regulation controlling which genes are active and thus shaping how cells develop and what goes wrong in certain conditions. Understanding this 3D architecture is paramount, and you might often run experiments to evaluate this in your research. But data analysis is key, including not just mastery with new tools but also knowing all relevant databases and applying the highest standards.

At Nexco we continuously thrive to stay at the vanguard of all our fields of action, keeping up with the latest resources and analytical approaches in frontiers bioinformatics and AI for life sciences, and publishing free posts that cover the most interesting advances. Here in this particular post we bring to you key new insights on how to more robustly decipher the complex spatial relationships at the core structure of the genome.

Recently, two important publications have significantly advanced our ability to both explore and reliably interpret the 3D genome. One published in Nucleic Acids Research introduces a comprehensive multi-omics database, EXPRESSO, that offers unprecedented access to diverse 3D genomic features. The other, a meticulous study in Nature Methods, provides an essential guide for comparing chromatin contact maps at scale, ensuring our interpretations are robust. Let’s delve into what these advancements mean for the future of genomics.

EXPRESSO: A Multi-omics gateway to the 3D genome

Published in Nucleic Acids Research as a “Breakthrough Article”, EXPRESSO (EXPloration of Regulatory Epigenome with Spatial and Sequence Observations) emerges as a powerful new database designed to illuminate the multi-layered 3D organization of the human genome. EXPRESSO (link to the article at the end) addresses several limitations of existing resources by:

• Integrating an impressive array of data, having combined 1360 3D genomic datasets (from Hi-C, HiChIP, ChIA-PET) with 842 1D genomic and transcriptomic datasets (from ChIP-seq, ATAC-seq, RNA-seq). Crucially, these come from the same biosamples across 46 different human tissues.
• Providing very rich annotations with diverse 3D genomic feature types, including A/B compartments, contact matrices, contact domains (TADs), architectural stripes, and chromatin loops.
• Boasting a user-friendly interface for data exploration and download, advanced visualization tools, and importantly, REST APIs for programmatic access — a boon for bioinformaticians. In fact, web-based applications within EXPRESSO allow users to directly correlate 3D genomic features with gene expression and epigenomic modifications seamlessly.

Some compelling examples

EXPRESSO’s power is showcased through compelling examples presented in the paper. For instance, it visually demonstrates how the WNT5a gene, implicated in prostate cancer, shifts from an inactive compartment in healthy prostate cells to an active compartment in cancer cells, alongside the formation of specific chromatin loops. Similarly, it reveals the formation of a “neo-contact domain” encompassing cardiac hypertrophy marker genes (NPPA, NPPB) in samples from patients with Dilated Cardiomyopathy, a structure absent in healthy hearts. These and other examples highlight how EXPRESSO can help researchers uncover tissue-specific and disease-associated alterations in 3D genome architecture.

By offering such a comprehensive and accessible platform, EXPRESSO aims to significantly deepen our understanding of how the genome’s spatial organization influences gene regulation and contributes to human health and disease.

Robustly comparing 3D genome maps

The second paper we bring to you here deals with how we can reliably compare chromatin contact maps to identify meaningful differences between cell types, conditions, or in response to perturbations. This very deep study was published in Nature Methods (link at the end of the post).

Numerous methods exist for comparing contact maps, but they often yield different results, and no “gold standard” has been established. To address this, the authors of this study meticulously evaluated 25 different ways to compare contact maps, using experimental Micro-C and Hi-C data as well as in silico-generated maps. Their findings were illuminating:

• First of all, it turned out that different methods prioritize different types of changes. Global comparison methods (like mean squared error) are useful for initial screening. However, to understand how/why maps diverge and to form specific functional hypotheses, biologically informed methods are necessary.
• Second, sensitivity is very important. The different methods vary in their robustness to biological and technical noise, changes in contact intensity versus structural patterns, or alterations like the gain/loss of CTCF-binding sites. Then, the study shows how, for example, Spearman correlation and Mean Squared Error can rank the same map pairs very differently based on their inherent sensitivities.

The study also introduces and validates four newly adapted contact map methods: Eigenvector, Contact Directionality, Distance Enrichment, and Triangle; and it demonstrates their utility and the often complementary nature to existing tools.

As a closing conclusion, the authors provide a reference guide, a codebase, and a thorough evaluation framework to help researchers select the most appropriate method(s) for their specific biological questions and data types.

We believe that this comprehensive benchmarking was very much needed in the field. Now, researchers to move beyond arbitrary choices of comparison metrics and instead make more informed decisions, leading to more reliable and reproducible insights into the dynamic nature of 3D genome organization.

Nexco’s perspective: integrating rich data with analytical rigor

The two advancements we have brought you here can be highly synergistic. EXPRESSO provides a rich, multi-layered dataset ripe for exploration, while the second study offers the methodological clarity needed to analyze such data — and indeed any 3D genomics data — with confidence.

At Nexco, our expertise lies precisely at this intersection. We understand that accessing and integrating complex multi-omics datasets, like those curated in EXPRESSO, is the first step. Our team is adept at navigating such resources and extracting relevant information for your specific research questions. Plus, of course applying the right analytical tools is crucial. Now, informed by rigorous evaluations like that in the Nature Methods study, we can help you choose or develop the most suitable computational pipelines to compare your chromatin contact maps, identify significant structural variations, and correlate these with functional genomic data.

Beyond just identifying differences, we help you understand their potential biological implications, whether in fundamental research or in the context of disease mechanisms and therapeutic development.

Want to explore the 3D architecture of your biological system or analyze complex chromatin contact data? Contact Nexco today, and let’s decode genome’s spatial secrets together.

Check in particular our 3D genomics services and our ChIP-seq services

References

• Cai et al (2025) EXPRESSO: a multi-omics database to explore multi-layered 3D genomic organization. Nucleic Acids Research 53(D1), D79–D90. https://academic.oup.com/nar/article/53/D1/D79/7875983
• Gjoni et al (2025) Comparing chromatin contact maps at scale: methods and insights. Nature Methods 22, 824–833. https://www.nature.com/articles/s41592-025-02630-5