Summary: For decades, textbooks described DNA as either “on” (unwrapped and active) or “off” (tightly wound around protein spools called nucleosomes). A new study overturns that binary view, revealing a far more nuanced system of regulation.
Using a new AI-driven computational method named IDLI, researchers found that nucleosomes behave like adjustable “volume dials” rather than fixed locks. By classifying 14 distinct nucleosome conformations, the team uncovered an organizational code that lets cells tune gene expression with precision.
Key Facts
- Beyond the Spool: Nucleosomes were long considered static obstacles to DNA access. Using IDLI (Iteratively Defined Lengths of Inaccessibility), researchers showed that more than 85% of nucleosomes are distorted or partially loosened, exposing segments of DNA.
- Fourteen Structural States: The study defines a “grammar” of 14 nucleosome shapes. Each state correlates with a different level of gene activity, making nucleosomes operate more like dimmer switches than binary on/off elements.
- AI-Powered Detection: IDLI analyzes two dimensions of data—across the DNA fiber and within individual nucleosomes—allowing the model to detect subtle internal distortions missed by previous methods.
- Transcription Factors as Sculptors: Transcription factors do more than locate open DNA. They actively reshape nucleosomes, pushing them into configurations that either permit or restrict gene reading.
- Implications for Disease and Aging: Many complex diseases and age-related changes may reflect misregulated gene expression levels rather than simple mutations. Mapping these 14 nucleosome states provides a new diagnostic and therapeutic framework.
Source: Gladstone Institute
Every human cell compacts over six feet of DNA into a microscopic nucleus. To package this long molecule efficiently and orderly, DNA wraps around octameric protein complexes known as nucleosomes.
For years scientists taught that DNA wrapped tightly around nucleosomes was inaccessible, and only unwrapped DNA could be transcribed. A collaborative study from Gladstone Institutes and the Arc Institute now challenges that simplistic model.
With an AI-assisted computational pipeline, the researchers discovered that most nucleosomes contain locally accessible sections of DNA—regions that are not fully wound and can be engaged by cellular machinery.
Published in Nature, the findings reveal a previously unrecognized layer of gene regulation rooted in the internal structure of nucleosomes.
“The old conception was that nucleosomes switch genes simply on or off. What we see instead is a spectrum—more like a volume dial,” says Gladstone Investigator Vijay Ramani, PhD, a lead author. “This represents an entirely new organizational code for chromatin.”
A More Detailed Readout of DNA Packaging
Although every cell contains the same genome, different cell types express different subsets of genes. Chromatin—the complex of DNA and nucleosomes—helps enforce that specificity by controlling access to genetic information. Scientists commonly analyze chromatin to infer which genes are active in a given cell.
The Ramani laboratory previously introduced SAMOSA, a method that maps nucleosome positions along single DNA molecules. IDLI builds on SAMOSA by applying an AI model to detect subtle structural differences within each mapped nucleosome footprint.
Instead of merely locating nucleosomes, IDLI inspects two-dimensional patterns: along the length of single chromatin fibers and within the wrap of each individual nucleosome. This allows the method to infer internal structural features and degrees of DNA exposure.
Chromatin Is Dynamic, Not Binary
A nucleosome comprises eight histone proteins; IDLI assesses whether all histone contacts are intact and tightly bound. When components are missing or loosened, the nucleosome is distorted and exposes patches of DNA.
Applying IDLI to mouse embryonic stem cell chromatin, the team found intranucleosomal accessibility in more than 85% of nucleosomes, indicating widespread structural variability.
“These results show the genome is far more dynamic and accessible than previously believed,” Ramani explains.
Importantly, the distortions were nonrandom and reproducible. The researchers categorized 14 distinct nucleosome states, each linked with different transcriptional outputs. The same patterns were detectable during human stem cell differentiation toward liver-like cells and in primary mouse liver tissue.
Arc Institute investigator Hani Goodarzi, PhD, who co-led the study, compares the advance to moving from a two-tone text to a full written language: “Chromatin isn’t just sound and silence anymore. We’re beginning to parse a nuanced grammar that controls gene expression.”
The study also shows that transcription factors directly influence nucleosome shapes. Depleting specific transcription factors altered distortion patterns predictably, indicating these proteins actively sculpt nucleosome architecture to regulate gene access.
“This expands the ways cells fine-tune gene activity—by making local stretches of DNA more or less accessible,” Ramani says.
Toward Better Understanding of Disease and Aging
Many complex diseases lack single-gene explanations because their root causes are distributed shifts in gene regulation. Small changes in expression levels across many genes—genes being partially active when they should be silent, or vice versa—can drive pathology.
Ramani views the 14 nucleosome states as measurable readouts of such shifts. “Complex diseases often involve gradations of activity,” he notes. “A gene might be expressed at half its normal level or active in the wrong cell type.”
IDLI also shows promise for aging research. Chromatin architecture follows predictable patterns of change with age, and some alterations appear reversible. Mapping nucleosome states across tissues and ages could reveal targets for restoring youthful gene regulation.
“We want to move beyond reading this chromatin language to learning how to write it—to intervene and restore healthy patterns,” Goodarzi adds.
About the Study
Funding: This work was supported by the National Institutes of Health, the California Institute for Regenerative Medicine, the Searle Scholars Program, and the W.M. Keck Foundation.
Key Questions Answered:
A: No. Partial accessibility is like a slightly open book: the gene is available to be read, but transcription still requires the right proteins to engage it. This partial exposure allows faster responses than fully unwrapping DNA.
A: Researchers generated long-read single-molecule chromatin maps with SAMOSA and trained an AI model to recognize specific accessibility signatures. Distorted nucleosomes show distinctive footprints—patterns of exposed DNA—that the model learns to identify.
A: That is an aspirational goal. Because chromatin structure shifts predictably with age, mapping healthy nucleosome grammars in young cells could guide therapies designed to restore youthful chromatin states in older cells.
Editorial Notes:
- This article was edited by a Neuroscience News editor.
- The journal paper was reviewed in full.
- Additional context was added by the editorial staff.
About this genetics research news
Author: Kelly Quigley
Source: Gladstone Institutes
Contact: Kelly Quigley – Gladstone Institutes
Image: Image credit: Neuroscience News
Original Research: Open access.
“Pervasive and programmed nucleosome distortion on single chromatin fibres” by Marty G. Yang, Hannah J. Richter, Simai Wang, Colin P. McNally, Camille M. Moore, Ali Emadi, Nicole E. Harris, Simaron Dhillon, Michela Maresca, Huimin Pan, Hayden Saunders, Ruiqiao Yang, Megan S. Ostrowski, Erika C. Anderson, Elzo de Wit, Jacquelyn J. Maher, Yuhong Fan, Geeta J. Narlikar, Elphège P. Nora, Holger Willenbring, Hani Goodarzi & Vijay Ramani. Nature
DOI:10.1038/s41586-026-10418-6
Abstract
Pervasive and programmed nucleosome distortion on single chromatin fibres
Despite decades of structural and biochemical studies, researchers have lacked a genome-scale method to assess variability in nucleosome structure along individual chromatin fibres. To fill that gap, the authors present Iteratively Defined Lengths of Inaccessibility (IDLI), a computational approach that maps single-molecule co-occupancy of distinct nucleosome and subnucleosome species alongside other protein–DNA interactions using long-read footprinting data.
IDLI classifies methylase-inaccessible footprints into categories including linker-histone-associated nucleosomes, nucleosomes with focal DNA accessibility across the wrap, unwrapped nucleosomes, and subnucleosomal species such as hexasomes and tetrasomes. Applying IDLI to mouse embryonic stem cell chromatin, the study finds that over 85% of nucleosomes contain intranucleosomal accessible DNA—what the authors term nucleosome “distortion.”
The analysis reveals domain- and expression-level-specific distortion patterns, including at promoters and satellite repeats. Occurrence of transcription factor motifs strongly correlates with particular distortion types, and degron-based perturbations provide evidence that transcription factors directly regulate these structural states. IDLI applied to in vitro human endoderm differentiation and to primary mouse hepatocytes shows distortion at pioneer factor FOXA2 binding sites, indicating that distortion is developmentally encoded and present in vivo. Genetic experiments in mice further demonstrate that a nucleosome-binding domain of FOXA2 affects nucleosome structure in vivo, implicating protein–nucleosome interactions as direct mediators of distortion.
Overall, the work documents extensive but regulated nucleosome structural variability at single-molecule resolution and offers a framework to model transcription factor binding, nucleosome remodeling, and cell-type-specific chromatin regulation across biological contexts.