Chromatin diminution, also termed programmed DNA elimination, removes chromosomal material during early development. In metazoans, this generally involves the loss of chromosomal fragments, predominantly heterochromatin, from cells of the somatic lineage. In protozoa, the process can remove entire chromosomes or sequences interspersed among genic loci.

The phenomenon is species-specific, and the diploid chromosome number after diminution may remain unchanged or change. Theodor Boveri first described chromatin diminution in 1887, and the review describes a research history extending for more than a century. During the fourth embryonic cleavage division, 94% of nuclear DNA is excised from the chromosomes of somatic-line cells, while germline cells retain their genome unchanged throughout development.

The diploid chromosome number remains unchanged, and the review states that the remaining 6% is sufficient for normal adult functions. Because this large fraction can be removed without disrupting somatic ontogeny, the authors treat the eliminated DNA as non-coding for the somatic lineage. The review positions C. kolensis as a model for investigating the C-value enigma: the large differences in genome size that do not correspond directly to differences in gene number or biological complexity.

Eukaryotic genome sizes can vary by more than 200,000-fold, while much of the excess DNA is non-coding. Proposed explanations have included selective neutrality, regulatory and structural functions, mutational pressure from insertions and deletions, and the possibility that genome-size changes are themselves subject to natural selection. The review presents C. kolensis as an experimental system for testing what happens when nearly all non-coding material is removed from a developing somatic genome.

Molecular analyses indicate that the eliminated material is not structurally simple. The review describes repeated sequences, spacers, mosaic fragments and direct and inverted repeats that are often highly similar or identical to consensus sequences. One eliminated repeat showed 97–98% sequence homology between Moscow and Baikal populations, with closely matching sequences retained in somatic cells.

The authors argue that this conservation is difficult to reconcile with a purely “junk” classification. They hypothesize that some eliminated sequences may have germline functions, including roles in meiosis or germ-cell maturation, but the review does not establish those functions directly. Chromatin diminution in C. kolensis is also associated with a lasting change in interphase-nuclear architecture.

Before diminution, embryonic cells have relatively homogeneous, diffuse nuclei. After the process, condensed heterochromatic segments and chromocenters become detectable, including chromocenters adjacent to the nuclear membrane, giving the nuclei an appearance resembling that of differentiated cells. The review interprets this transition as coinciding with the maternal-to-zygotic transition and as evidence that diminution may participate in the regulation of differentiation between somatic and germ lineages.

Timing experiments provide additional evidence for nuclear control of the process. In a cited irradiation experiment, high radiation doses that impaired nuclear-genome function stopped chromatin diminution. The authors infer that the mechanisms regulating the morphogenetic function of C. kolensis nuclei are triggered during the fourth cleavage division, when chromatin reduction occurs.

This finding supports initiation through zygotic nuclear activity, although the review does not identify the individual regulatory factors or establish the full molecular pathway. Experiments in C. kolensis also found that chromosome-aberration frequencies were 30–50 times lower after diminution than before it. That change was larger than the roughly 15- to 16-fold reduction predicted if aberration frequency simply scaled with the amount of remaining DNA.

The review discusses a model in which aberrations arise disproportionately in DNA associated with nuclear-matrix attachment regions and diminution removes a disproportionate share of those regions. The proposed reduction in DNA–matrix contact points remains an assumption in this explanatory model rather than a directly demonstrated result. The review also considers how non-coding DNA might influence recombination and genome evolution.

One hypothesis proposes that non-coding sequences separate genes, exons, introns, regulatory elements and structural features, thereby increasing opportunities for recombination and generating a wider range of sequence variants. Under this model, genome reduction after adaptation to a narrow ecological niche could limit further exploration of sequence space. A separate evolutionary model places chromatin diminution within incomplete genome reduction: somatic cells shed DNA that is unnecessary for later development, whereas germline cells retain sequences needed for meiosis and early embryogenesis.

Possible reproductive isolation is also examined. Russian and German C. kolensis populations differ in the timing of diminution and in the distribution of eliminated chromatin during anaphase, while German Cyclops insignis reportedly undergo diminution whereas Russian populations do not. Models combining genome reduction, altered chromosome order and hybrid meiosis propose a route to reproductive barriers.

These scenarios are hypothetical, and the review does not demonstrate that chromatin diminution caused speciation.

Subject of Research: Review of programmed DNA elimination, genome reduction and cellular differentiation in animals

Article Title: Chromatin diminution as a tool to study some biological problems

Article References: Grishanin, A. (2024). Chromatin diminution as a tool to study some biological problems. Comparative Cytogenetics, 18, 27-49. https://doi.org/10.3897/compcytogen.17.112152

Image Credits: AI Generated

DOI: 10.3897/compcytogen.17.112152

Keywords: chromatin diminution, programmed DNA elimination, genome reduction, non-coding DNA, copepods, Cyclops kolensis, nuclear architecture, cell differentiation, genetic isolation, C-value enigma, Chromatin, diminution

The study was co-led by Dr Maria Carolina Florian, a researcher in the Regenerative Medicine program at IDIBELL and ICREA Research Professor, and Dr Paula Petrone, a researcher at BSC-CNS and ISGlobal, a centre supported by the la Caixa Foundation. The work formed a central part of the doctoral thesis of Pablo Iañez, a researcher at ISGlobal, and combined expertise in stem cell biology, aging, image analysis and machine learning. Their starting point was a well-known clinical and biological problem: as the body ages, the hematopoietic system, the network of organs and tissues responsible for producing blood cells, progressively loses its capacity to maintain adequate blood cell production. Measuring this decline at the level of individual stem cells is essential for understanding hematopoietic aging and for testing strategies designed to slow or reverse it.

Chromatin is the material that makes this measurement possible. It consists mainly of DNA wrapped around proteins, and its spatial organization within the nucleus regulates gene activity, ultimately determining cell identity and function. As stem cells age, this organization changes, but the changes are subtle, heterogeneous and often invisible to a trained observer looking through a microscope. That is precisely where the new model excels. ChromAgeNet is a convolutional neural network, a class of artificial intelligence model designed to analyze images, and it was trained to distinguish young from aged hematopoietic stem cells based solely on the appearance of the nucleus.

To build and train the model, the researchers used three-dimensional microscopy images of mouse hematopoietic stem cell nuclei stained with DAPI, a simple, inexpensive and widely used fluorescent dye that binds to DNA and reveals the distribution of chromatin density inside the nucleus. The choice of DAPI is significant for practical reasons. Because it is a low-cost stain that can be incorporated easily into existing microscopy protocols, the approach does not require exotic reagents or specialized sample preparation, which makes it realistic to deploy in high-throughput microscopy workflows where thousands or millions of samples must be processed automatically.

The performance results are striking. Based on the appearance of the nucleus alone, ChromAgeNet demonstrated a 77 percent probability of correctly sorting cells into two groups, young versus aged. Remarkably, this exceeded the performance of a machine learning model built on chromatin features that the researchers themselves had previously defined by hand. In other words, the deep learning system extracted predictive information from the images that went beyond the features human experts had considered informative. This is a recurring theme in modern image-based biology: neural networks can detect combinations of visual cues that no single engineered feature captures, and here those cues turned out to encode biological age.

One of the most important aspects of the work is that the aging-associated differences are not necessarily visible to the naked eye in the microscopic images. Stem cell aging is a heterogeneous process, meaning that different cells age in different ways and to different degrees, and changes in nuclear architecture can be extremely subtle. ChromAgeNet nevertheless manages to identify combinations of spatial chromatin features that carry information about the aging state of individual cells. This transforms a subjective, qualitative impression of an aged nucleus into a quantitative, scalable measurement that can be applied consistently across large cell populations.

Crucially, the team did not stop at classification. They analyzed the model’s behaviour to determine which specific features of the images were most helpful in distinguishing young cells from aged cells. This interpretability analysis identified several predictive features of the age-associated state, including chromatin entropy, a measure of the disorder or variability in chromatin organization; heterochromatin, the densely packed and generally gene-poor form of chromatin, located at the periphery of the nucleus; and certain chromatin condensates, compact clusters of the DNA-protein complex. Knowing which elements of DNA organization the model relies on allows researchers to go beyond a simple young-versus-old label and gain mechanistic insight into the nuclear architecture of aging.

This mechanistic dimension positions ChromAgeNet as a complement to existing biological age biomarkers rather than a replacement for them. The best known of these are the so-called epigenetic clocks, which estimate biological age from chemical modifications to DNA, particularly methylation patterns. Epigenetic clocks require sequencing-based measurements, whereas ChromAgeNet requires only a stained image, so the two approaches interrogate different layers of the aging process. Combining image-derived chromatin architecture with molecular biomarkers could yield a richer, multi-dimensional picture of how a given stem cell is aging and how it responds to experimental interventions.

The most forward-looking application explored in the study is screening for rejuvenation strategies. As a proof of concept, the researchers applied ChromAgeNet to aged hematopoietic stem cells that had been treated with different epigenetic drugs, asking whether the treatments produced changes in chromatin organization compatible with a younger nuclear state. The authors are careful about what this does and does not show. The results do not demonstrate that the treatments functionally rejuvenated the cells, but they do establish the potential of ChromAgeNet as a tool to detect age-associated chromatin changes in response to interventions. That distinction matters: a model that can flag candidate compounds for closer functional testing accelerates discovery without overclaiming therapeutic effect.

The combination of cheap staining, a small number of model parameters and compatibility with high-throughput microscopy makes the platform unusually scalable. In principle, the same pipeline could be used to study large libraries of compounds, rapidly identifying those that shift the nuclear architecture of aged stem cells toward a younger configuration, which would then be prioritized for deeper functional assays. To support exactly this kind of community-driven development, the researchers have released a dataset of three-dimensional images of hematopoietic stem cells together with ChromAgeNet itself. This is a meaningful contribution in its own right, because publicly available imaging datasets for hematopoietic stem cell aging are currently scarce, and shared benchmarks are what allow computational tools to be validated and improved across laboratories.

Taken together, the study demonstrates that the three-dimensional organization of DNA contains quantifiable information about the aging of blood stem cells, and that artificial intelligence can extract that information from routine microscopic images. ChromAgeNet provides a new, faster and more scalable way to study cellular aging and to explore strategies for preserving or recovering hematopoietic stem cell function. If the approach generalizes beyond mouse cells and finds adoption in screening pipelines, the humble DAPI-stained nucleus, an image type generated in cytometry and microscopy labs every day, may become one of the most informative and inexpensive windows into the biological age of the cells that keep our blood, and ultimately our bodies, running.

Subject of Research: Deep learning detection of hematopoietic stem cell aging from 3D chromatin architecture in nuclear images

Article Title: AI detects signs of aging in blood stem cells from nuclear images

Article References: AI detects signs of aging in blood stem cells from nuclear images. (n.d.). Original publication

Image Credits: AI Generated

DOI: Not provided

Keywords: artificial intelligence, chromatin, hematopoietic stem cells, aging, DAPI staining, convolutional neural network, nuclear architecture, epigenetic drugs, rejuvenation, Aging Cell, biomarkers, high-throughput microscopy

The stakes of this research are enormous. Genome organization is intimately connected to essential cellular functions such as gene expression, DNA replication, and DNA repair. Enhancers, the distal regulatory elements that boost gene activity, may lie millions of DNA letters away from the genes they control along the linear chromosome, yet they must physically find and contact their targets within the crowded nuclear space. Understanding how the chromatin fiber brings the right pieces together requires more than biology alone; it requires quantitative models in which chromosomes are treated as polymer chains and proteins as interacting particles whose collective behavior can be simulated, perturbed, and compared against experimental data from microscopy and chromosome-conformation assays.

The first pillar of the framework discussed by Fontana and Chiariello is phase separation, the same process that causes oil to bead up in water, here operating on proteins inside the nucleus. Many regulatory proteins carry intrinsically disordered regions that allow them to interact with copies of themselves, forming stable condensates. Well-studied examples include Mediator and RNA polymerase II, which cluster together at active genes, as well as coactivators that assemble at super-enhancers and transcription factors whose activation domains can condense to drive gene expression. At larger scales, phase separation helps build Cajal bodies and the nucleolus, a multi-phase structure assembled from distinct molecular species of proteins and RNA. Crucially, when the phase-separation properties of these proteins go wrong, the result can be disease, ranging from rare genetic disorders involving nucleolar dysfunction to cancer driven by aberrant condensates.

To capture this phenomenon computationally, theorists have built models in which chromatin is represented as a polymer and proteins are represented as particles that bind to it. Some of the simplest versions include only polymer-protein interactions, leaving out direct protein-protein attraction. Even so, these bridging models give rise to what is known as polymer-polymer phase separation, in which the polymer itself nucleates condensates of bound proteins that then glue distant chromatin segments together. Despite their minimalism, such models have successfully reconstructed the three-dimensional structures of real genomic regions, described chromatin remodeling across different cell types, and even tracked how genome architecture is rearranged following viral infections, including SARS-CoV-2.

More realistic systems require explicit protein-protein interactions alongside chromatin-protein affinity, and, in the most complete versions, loop extrusion as well. In loop extrusion, molecular motors of the cohesin or condensin families slide along the chromatin fiber, enlarging DNA loops until they are halted at anchor points, often corresponding to binding sites of the architectural protein CTCF. Single-molecule imaging has now captured this process in real time, and computational studies show that extrusion contributes to global chromosome condensation during mitosis as well as to the local loops and domains that organize the interphase genome. When all these ingredients coexist, the behavior of the system is governed by a set of control parameters, including protein concentration, self-interaction affinity, chromatin affinity, and motor density and processivity, and the resulting equilibrium states fall into complex phase diagrams reminiscent of those studied in condensed matter physics.

One of the most striking predictions to emerge from large-scale molecular dynamics simulations concerns a scenario now called polymer-assisted condensation, or PAC. Simulations revealed that phase-separated condensates can arise in two distinct ways: either spontaneously through strong protein-protein affinity, independently of the polymer, or in a polymer-assisted route in which protein-protein attraction alone is too weak to drive condensation, but binding to chromatin concentrates the proteins enough to tip the system over the threshold. This second mechanism, first predicted computationally and later explored theoretically, has been proposed as a potentially general principle of chromatin organization, with implications for heterochromatin formation and epigenetic memory.

The power of the integrated approach was demonstrated in recent work on PHF13, a protein that binds chromatin enriched with the histone methylation marks H3K4me3 and H3K27me3 and can also self-associate through specific N-terminal sub-domains. Molecular dynamics simulations showed that when both chromatin-protein and protein-protein interactions were enabled at sufficient PHF13 concentration, the modeled polymer collapsed into a compact, roughly spherical globule decorated with PHF13 clusters, mirroring microscopy images of interphase cells overexpressing PHF13, in which chromosomes condense into rounded structures. When the self-interaction domains were disabled in silico, compaction was reduced and condensation disappeared, exactly matching the behavior of engineered PHF13 variants lacking those domains in living cells. The close agreement between simulation and experiment provides a compelling validation of the physics-based framework.

The same modeling philosophy is now being applied to disease. The chimeric protein NUP98-HOXA9, an oncofusion associated with leukemia, can undergo phase separation and bind chromatin, promoting the formation of aberrant loops that bring oncogenes such as PBX3 into contact with distant regulatory elements. Models of the relevant genomic regions accurately recapitulated the aberrant looping patterns observed in experimental Hi-C contact maps from human cells and in single-molecule mobility measurements, while also capturing the interplay between phase separation and loop extrusion that underlies the pathological rewiring. Because altered phase separation has been implicated in cancers and rare genetic diseases, such models offer a route to understanding how molecular-scale misfolding translates into gene misregulation.

The third pillar of the framework concerns the nuclear envelope. The nuclear lamina, a protein network embedded in the inner membrane, adsorbs peripheral chromatin, spatially segregating specific genomic regions from the nuclear interior and repressing their activity. Computational models represent the lamina as an adsorbing membrane that attracts or sterically excludes chromatin segments, with interactions parameterized from DamID profiles that quantify locus-specific lamina association. When loop extrusion and lamina adsorption are combined in a single simulation, the framework accurately reproduces the organization of peripheral loci such as the mouse Sox9 and Kcnj2 region, and tuning chromatin-lamina affinity allows different degrees of peripheral association and distinct cell populations to be described. Simulations of genomic editing, including deletion of CTCF boundary elements, quantitatively matched experimental outcomes, establishing a link between loop extrusion and lamina association even though the two processes act independently.

Fontana and Chiariello conclude that the great advantage of these physics-based models is their versatility: phase separation, loop extrusion, and membrane adsorption can be combined within a single computational framework to describe increasingly detailed biological contexts, at the cost of larger parameter spaces and more complex phase diagrams that molecular dynamics simulations are well suited to explore. Looking forward, the authors point to promising extensions, including modeling pathogenic protein variants, the chromatin changes of ageing, and the architectural disruptions induced by external pathogens such as viruses. As genome-wide datasets grow richer and simulations grow faster, the dream of predicting, and perhaps one day correcting, the physical misfolding events behind human disease moves steadily closer to reality.

Subject of Research: Physics-based computational modeling of chromatin 3D organization through phase separation, loop extrusion, and nuclear lamina adsorption

Article Title: On the physics of chromatin molecular complex: from phase-separation to membrane adsorption

Article References: Fontana, A., & Chiariello, A. M. (2026). On the physics of chromatin molecular complex: from phase-separation to membrane adsorption. Frontiers in Molecular Biosciences, 13, Article 1951961. https://doi.org/10.3389/fmolb.2026.1951961

Image Credits: AI Generated

DOI: 10.3389/fmolb.2026.1951961

Keywords: chromatin, 3D genome organization, phase separation, loop extrusion, polymer physics, nuclear lamina, molecular dynamics, gene regulation, biomolecular condensates, CTCF, cohesin, leukemia

The research hinged on a technical feat: watching individual nucleosomes, the fundamental repeating units of chromatin, move in real time inside living cells. The team combined single-nucleosome imaging and tracking with super-resolution three-dimensional structured illumination microscopy, or 3D-SIM, which pushes past the classical diffraction limit of light microscopy to resolve chromatin organization at a scale of tens of nanometers. Together these methods allowed the investigators to measure both the local motion of nucleosomes and the three-dimensional morphology of condensed euchromatic domains in the same living cells, before and after acute manipulation of cohesin.

Euchromatin is not a homogeneous soup of DNA and proteins. It is partitioned into domains with distinct physical properties, some of which behave like condensed droplets enriched in transcriptionally active machinery. A central question in the field has been whether these domains are held apart by active mechanisms or whether they self-organize through the physics of phase separation, with weak, multivalent interactions driving chromatin to demix into distinct liquid-like compartments. The new findings point to a decisive role for cohesin in maintaining this compartmentalization, adding a genetic and biochemical handle to what has often been framed as a purely biophysical problem.

The single-nucleosome tracking experiments provided a dynamic readout of chromatin fluidity. By labeling histones sparsely and following their trajectories frame by frame, the researchers could quantify how freely nucleosomes diffuse within their local chromatin environment. In normal cells, nucleosomes within condensed euchromatic domains showed constrained mobility, consistent with a densely packed and relatively stable chromatin fiber. When the investigators acutely depleted cohesin, the picture changed dramatically: nucleosome mobility increased, indicating that the condensed euchromatic domains had become more fluid, more like a liquid and less like a restrained polymer network.

What makes this result particularly striking is what did not change. Overall chromatin compaction, measured as the average density of chromatin within the euchromatic regions, was essentially unaffected by the loss of cohesin. The domains remained condensed; they did not dissolve or decondense. Instead, their internal fluidity rose, and with it their tendency to coalesce. In other words, cohesin does not maintain the physical state of euchromatin by controlling how tightly it is packed. Its role is spatial rather than material: it keeps condensed domains in their proper places, preventing neighboring territories from merging even while each territory stays as compact as before.

Super-resolution 3D-SIM imaging made this loss of spatial control visible. In cohesin-proficient cells, condensed euchromatic domains appeared as discrete, bounded objects distributed through the nuclear interior. After cohesin removal, the boundaries between adjacent domains blurred, and domains that had previously remained separate were observed to fuse locally, producing larger, mixed territories. The consequence of this mixing is not merely cosmetic. The functional identity of chromatin domains depends on their insulation: enhancers, silencers, and other regulatory elements act over defined genomic distances, and the three-dimensional organization of chromatin helps enforce those limits. When domains mix, regulatory inputs can reach genes that were previously shielded from them.

That functional consequence was confirmed at the level of transcription. The researchers found that cohesin loss compromised transcriptional insulation, allowing the transcriptional programs of neighboring domains to interfere with one another. Genes that had been maintained in distinct regulatory environments began to respond to the wrong controls. This connects the physical observation, domain mixing, to a biological output, misregulated gene expression, and suggests that the barrier function of cohesin is not an incidental byproduct of loop extrusion but a property that cells rely on to keep their regulatory circuits orderly.

Cohesin is already famous for its role in genome architecture through the process of loop extrusion, in which the complex translocates along DNA and enlarges loops until halted by boundary elements such as CTCF. This activity organizes the genome into topologically associating domains, or TADs, and its disruption in diseases known as cohesinopathies causes developmental defects. The new work adds a second, complementary dimension to this picture. Beyond extruding loops between distant genomic sites, cohesin locally constrains condensed euchromatin, and these two functions together may explain why mutations affecting cohesin produce such broad and pleiotropic effects on gene expression during development.

The findings also sharpen the ongoing debate about how chromatin compartments form and are maintained. Models based on phase separation predict that chromatin domains with similar biochemical properties should spontaneously coalesce and repel dissimilar ones, with demixing driven by the intrinsic interactions among nucleosomes, histone tails, and associated factors. If phase separation alone were sufficient to maintain euchromatic domains, removing cohesin should not have caused them to merge; their physical similarity should have kept them together but distinct. The observation that cohesin loss increases fluidity and permits mixing suggests that intrinsic chromatin interactions favor coalescence, and that cohesin actively opposes this tendency, fencing off domains that would otherwise blend. Cohesin, in this view, is the factor that prevents local mixing of condensed euchromatin in the crowded nuclear environment.

Technically, the study demonstrates the power of imaging-based approaches that operate in living cells at single-molecule and super-resolution scales simultaneously. Fixed-cell methods capture snapshots of architecture but cannot distinguish between domains that are statically separate and domains that are dynamically prevented from mixing. Single-nucleosome tracking adds the temporal dimension, revealing that cohesin constrains motion rather than compaction, a distinction invisible to purely structural measurements. The combination of these modalities provides a template for future studies of how other nuclear factors, from architectural proteins to RNA polymerase machinery, shape the physical behavior of chromatin in real time.

The broader implications reach into human disease. Cohesin genes are among the most frequently mutated in cancers, and cohesion-related developmental syndromes such as Cornelia de Lange syndrome arise from haploinsufficiency of cohesin components. If part of the pathology of these conditions stems from the loss of transcriptional insulation, then the specific mechanism identified here, increased fluidity and mixing of condensed euchromatic domains, offers a concrete physical picture of how cohesin mutations dysregulate gene expression. It also suggests that the degree of chromatin fluidity could serve as a measurable cellular phenotype, detectable by live-cell imaging, in cells carrying cohesin mutations or in tumors with disrupted cohesin function.

Looking forward, the study raises questions that will drive the next phase of research. How does cohesin physically constrain the mobility of condensed euchromatin, and is this barrier function mediated by the same extrusion activity that forms loops, or by a distinct, stabilizing interaction with chromatin? How many cohesin complexes are required to fence off a single condensed domain, and where do these barrier-forming complexes reside relative to domain boundaries? And do other nuclear factors cooperate with cohesin in maintaining compartmental identity, or does cohesin act largely alone? Answering these questions will require extending the same live-cell, super-resolution toolkit used in this study, but the central message is already clear: the genome’s active regions are kept separate not by the physics of chromatin alone, but by an active, energy-consuming molecular machine that patrols the boundaries between them.

Subject of Research: Cohesin-mediated spatial insulation of condensed euchromatic domains in living human cells

Article Title: Cohesin prevents local mixing of condensed euchromatic domains in living human cells

Article References: Shimazoe, M. A., Iida, S., Minami, K., Higashi, K., Tamura, S., Kobayashi, Y., Fujishiro, S., Xiong, L., Nakazato, K., Ashwin, S. S., Nishiyama, T., Nagata, Y., Kanemaki, M. T., Kawaguchi, A., Ohkawa, Y., Schermelleh, L., Toyoda, A., Xie, L., Kurokawa, K., … Maeshima, K. (2026). Cohesin prevents local mixing of condensed euchromatic domains in living human cells. Nature Genetics, 58(9), 2335-2349. https://doi.org/10.1038/s41588-026-02736-2

Image Credits: AI Generated

DOI: 10.1038/s41588-026-02736-2

Keywords: cohesin, chromatin, euchromatin, single-nucleosome imaging, super-resolution microscopy, 3D-SIM, genome organization, transcriptional insulation, nuclear architecture, phase separation, loop extrusion, epigenetics