A team led by Professor Inkyung Jung from the Department of Biological Sciences at KAIST, working with Professor Yarui Diao’s group, has published a new single-cell method called scHiCAR that simultaneously captures a cell’s active genes, its open regulatory regions, and the physical folding of its DNA. Applied to 1.62 million mouse brain cells, the technique resolves 22 distinct cell types and maps enhancer-to-promoter contacts at 5-kilobase resolution, offering a level of combined biological detail that no prior method has matched at this scale. The work, reported in Nature Biotechnology, arrives as neuroscience and disease research increasingly demand tools that can link gene regulation to three-dimensional genome architecture inside individual cells.
Three Layers of Biology From One Cell
Most single-cell genomics tools measure one or two molecular layers at a time: gene expression, chromatin accessibility, or DNA contacts. scHiCAR captures all three. It is a plate-based combinatorial barcoding method that reads mRNA for the transcriptome, open chromatin for the epigenome, and chromatin contacts for the 3D genome from the same individual cells. That triple readout matters because knowing which genes are turned on, which regulatory switches are accessible, and how distant DNA elements physically loop together gives researchers a far more complete picture of how a cell controls its own identity.
The three-dimensional organization of cis-regulatory elements is central to transcription control, as the study’s authors emphasize in their analysis of genome folding. A gene may sit thousands of base pairs away from the enhancer that activates it; only by measuring the physical contact between the two can scientists confirm that the regulatory connection actually exists in a given cell type. Previous approaches forced researchers to infer these links by overlaying data from separate experiments run on different cell populations, introducing noise and losing single-cell resolution. scHiCAR eliminates that workaround by generating all three data types from one nucleus, allowing direct assignment of regulatory loops, chromatin accessibility, and transcriptional output within the same cell.
Scale, Cost, and What 1.62 Million Cells Reveal
Raw throughput is where scHiCAR separates itself from earlier tri-omic attempts. The team profiled 1.62 million mouse brain cells and classified them into 22 brain cell types, then identified enhancer-promoter pairs at 5-kilobase resolution across those populations. That granularity is fine enough to distinguish regulatory contacts between neighboring genes, a task that bulk Hi-C experiments typically blur. The per-cell cost sits at approximately $0.04, according to KAIST’s institutional release, though no peer-reviewed economic breakdown has been published to date. If that figure holds under independent replication, it would place scHiCAR’s pricing well within reach of large-scale atlas projects that routinely process hundreds of thousands of cells to chart cellular diversity and regulatory wiring.
Validation experiments were run in both human H1 embryonic stem cells and mouse embryonic stem cells, with raw data deposited in the Gene Expression Omnibus under accession GSE267117. Public availability of the underlying tar files and contact matrices is a practical step that lets other labs reproduce and extend the work, though the dataset currently covers cell lines rather than primary human tissue. That distinction is important: cell-line validation confirms that the chemistry and barcoding strategy function as intended, but demonstrating consistent performance in freshly dissected human brain samples or patient-derived organoids remains an open task that will determine how broadly scHiCAR can be deployed in translational settings.
How scHiCAR Fits the Single-Cell 3D Genome Toolkit
The field has been converging on multi-modal single-cell assays for several years, and scHiCAR builds directly on that trajectory. An earlier method called sn-m3C-seq demonstrated joint measurement of DNA methylation and chromatin conformation in single human cells, pairing an epigenomic layer with 3D genome data but omitting transcription; that work, described in a Nature Methods article, showed that combinatorial barcoding could scale single-cell chromatin contact maps into the tens of thousands of cells. More recently, droplet-based approaches achieved scalable single-cell Hi-C profiling of chromatin architecture in heterogeneous tissues, with one platform in Nature Biotechnology extending high-throughput 3D genome mapping to complex organs and disease samples. A separate droplet method paired single-cell Hi-C with RNA profiling for combined 3D genome and transcriptome readouts, as detailed in a cell research study that linked chromatin loops to gene activity in individual cells.
A tri-omic mapping study published in Nature Methods integrated chromatin accessibility, chromatin interactions, and transcription in whole mouse brains across the lifespan, providing a proof-of-principle that three modalities could be captured together at single-cell resolution. That work, reported in a lifespan brain atlas, demonstrated that developmental and aging trajectories can be traced through coordinated changes in 3D genome structure and regulatory element activity. scHiCAR’s contribution is not the conceptual leap of triple measurement itself but the combination of throughput, cost, and resolution it achieves in a plate-based format. Processing more than 1.6 million cells at roughly four cents each while resolving 5-kilobase enhancer–promoter contacts represents a practical jump from proof-of-concept to production-scale utility. At the same time, head-to-head benchmarks comparing sensitivity, false-discovery rates, and per-modality sequencing depth across these competing platforms have not yet been published, leaving open questions about where scHiCAR’s chemistry may trade depth for breadth in large experiments.
Why Tri-Omic Data Changes the Research Calculus
For neuroscientists studying disorders like Alzheimer’s disease or autism, the ability to see which enhancers physically contact which promoters in a specific neuronal subtype can change how hypotheses are framed. Genome-wide association studies often implicate noncoding variants that fall in enhancers rather than in protein-coding exons, making it difficult to infer which genes are actually affected. With a tri-omic map, researchers can ask whether a variant-bearing enhancer is accessible in a given cell type, whether it forms a 3D contact with a nearby promoter, and whether that promoter’s gene is transcriptionally active in the same cells. That combination of accessibility, contact, and expression turns a static list of risk loci into a set of testable mechanistic models about disrupted regulatory loops in defined neural populations.
The same logic extends to developmental biology and cancer research. In embryonic tissues, cell fate decisions are orchestrated by waves of enhancer activation and chromatin reorganization; tri-omic data can reveal whether a failure to adopt a particular lineage reflects missing enhancer accessibility, absent 3D contacts, or downstream transcriptional blocks. In tumors, where chromatin topology is frequently rewired, scHiCAR-style maps could expose oncogenic enhancer hijacking events that bring powerful regulatory elements into contact with proto-oncogenes in only a subset of malignant cells. Because all three layers are measured in the same nucleus, rare but consequential regulatory configurations are less likely to be averaged away, enabling more precise targeting of the cellular subclones that drive disease progression or therapy resistance.
More from Morning Overview
*This article was researched with the help of AI, with human editors creating the final content.
