The review, authored by Zhe Li, Zhaosong Meng, Lei Sui, Chufan Ma, and colleagues from institutions including The Fourth Military Medical University and Tianjin Medical University, advances what the authors describe as an evidence-weighted, locus- and complex-centered framework. Rather than treating MED1 as simply “pro-” or “anti-tumor,” the authors distinguish between causal genetic and mechanistic data on one hand and clinicopathological correlations on the other. This distinction matters, they argue, because much of the confusion in the MED1 literature stems from studies that observed associations between MED1 abundance and patient outcomes without establishing whether MED1 actually drives the biology in question.

MED1, also known as TRAP220 or PBP, is a non-DNA-binding transcriptional co-regulator. It does not bind DNA directly, and it does not catalyze any reaction. Instead, it functions as part of the massive Mediator complex, a multi-protein assembly that physically connects DNA-bound transcription factors to RNA polymerase II, the enzyme that reads genes and produces messenger RNA. MED1 is particularly interesting within this complex because it contains domains that interact with nuclear receptors, the family of ligand-activated transcription factors that includes the estrogen receptor and the androgen receptor, two of the most clinically important drivers of hormone-responsive cancers.

The strength of direct mechanistic evidence varies dramatically across cancer types, and the review maps this landscape carefully. In estrogen receptor-driven breast cancer, the evidence that MED1 supports oncogenic transcription is robust. MED1 is recruited to estrogen receptor-bound enhancers, where it helps deploy the transcriptional machinery needed to activate genes promoting proliferation and survival. Similarly, in androgen receptor-driven prostate cancer, MED1 contributes to the expression of the androgen receptor’s target gene program. The authors also point to E2A-PBX1-positive B-cell acute lymphoblastic leukemia and hepatocyte tumor models as contexts where direct experimental manipulation of MED1 has demonstrated its oncogenic function.

But the picture inverts in other settings. In defined models of non-small-cell lung cancer and melanoma, the loss of MED1 promotes invasive behavior rather than suppressing it. This is a striking finding because it suggests that MED1 can act as a tumor suppressor in certain cellular contexts, restraining the transcriptional programs that drive invasion and metastasis. The review also notes that findings in colorectal and bladder cancer remain largely correlative, meaning that while MED1 expression patterns may track with clinical features, causality has not been established through mechanistic experiments.

What determines which way MED1 tips in any given cancer? The review identifies several interacting determinants. Lineage-specific transcription-factor recruitment plays a central role: which transcription factors are present and active in a particular cell type determines which genomic loci MED1 occupies and which gene programs it influences. Signaling-dependent modification also matters, since MED1 is subject to post-translational modifications that alter its behavior in response to external signals. Chromatin state and broader cellular context further shape MED1’s output, meaning that the same protein operating on different genomic terrain can produce entirely different consequences for the cell.

One of the more technically interesting sections of the review addresses MED1’s role in super-enhancers and transcriptional condensates. Super-enhancers are dense clusters of enhancer elements that drive high expression of genes critical for cell identity, and they are often marked by unusually high concentrations of transcriptional machinery, including Mediator complex components. Some research has suggested that these regions form phase-separated condensates, membrane-less compartments that concentrate transcriptional regulators. MED1 has been reported as enriched at these assemblies, and its intrinsically disordered regions have been implicated in condensate formation. The review, however, sounds a note of caution: enrichment of MED1 at super-enhancer-associated structures, while real, does not by itself establish that MED1 is structurally or functionally necessary for condensate integrity or function. This distinction is important because the field has sometimes moved quickly from observing co-localization to inferring dependency, and the authors argue for more rigorous perturbation-based evidence before drawing such conclusions.

The therapeutic implications of this framework are significant but tempered. On one hand, the finding that MED1 supports oncogenic transcription in breast and prostate cancer suggests that disrupting MED1-dependent complexes could be a powerful therapeutic strategy, particularly because it might undercut multiple oncogenic programs simultaneously rather than targeting a single signaling pathway. On the other hand, the review notes that no clinically validated MED1-selective inhibitor or degrader currently exists. Pharmacologic strategies aimed at MED1 have so far been indirect, targeting the proteins it partners with or the signaling pathways that activate it, while RNA-based suppression approaches remain largely experimental. The challenge is compounded by MED1’s role as a co-regulator rather than an enzyme: it lacks the catalytic pockets that make many cancer drug targets tractable, and any therapeutic strategy must contend with the risk of disrupting its physiological functions in normal tissue.

This is where the context-dependent framework becomes more than an academic exercise. By shifting the focus from pan-cancer expression patterns to partner-, locus-, and model-specific dependency, the authors define testable biomarkers and therapeutic hypotheses. Rather than asking whether MED1 is high or low in a given tumor, clinicians and researchers could ask which transcription factors MED1 is partnered with, which loci it occupies, and whether the tumor’s survival depends on MED1-containing complexes at specific oncogenic super-enhancers. This approach preserves the possibility of selectively targeting oncogenic MED1 complexes while sparing the physiological MED1 functions that normal cells rely on, and it explains why MED1 loss might be harmful in some cancers and helpful in others.

The review also highlights how the MED1 paradox illustrates a broader lesson for cancer biology. Transcriptional co-regulators occupy an awkward middle ground in the oncogene-versus-tumor-suppressor taxonomy. Because their function is entirely dependent on the transcription factors they serve and the genomic context in which they operate, their role in disease is inherently contextual. This means that large-scale correlative studies, however well powered, will continue to produce contradictory results unless they are designed to capture the mechanistic variables that actually determine function. The authors’ framework, grounded in mechanistic evidence from breast cancer, prostate cancer, leukemia, hepatocyte models, lung cancer, and melanoma, offers a template for how to approach other co-regulators whose roles have been similarly contested.

As research moves forward, several questions stand out. Can MED1-selective degraders be developed using emerging protein degradation technologies? Are there specific super-enhancer contexts where MED1 dependency is absolute, offering a therapeutic window? And can the lineage-specific determinants of MED1’s function be mapped systematically across cancer types to produce a predictive atlas? The answers will determine whether MED1 transitions from a molecule of biological intrigue to a clinically actionable target. For now, the review makes a compelling case that the answer to “what does MED1 do in cancer?” is neither “one thing” nor “it depends on nothing,” but rather a precise, mechanistically grounded dependence on partners, loci, and context that can, in principle, be measured, modeled, and ultimately exploited.

Gene regulation fundamentally depends on the orchestrated co-binding of proteins across chromosomes. These proteins, including transcription factors, chromatin remodelers, and structural components, interact dynamically along chromatin fibers, determining how genes are expressed in specific cells. Yet, until now, current methodologies have fallen short in deciphering the heterogeneity in regulatory protein binding between homologous chromosomes or even between individual cells. What emerges is a blurred picture, averaging signals over populations and masking critical functional nuances necessary for understanding disease processes and cellular differentiation. DAF-seq surmounts these challenges by combining DNA sequence information with precise protein occupancy fingerprints on individual chromatin fibers.

At its core, the DAF-seq technique harnesses deaminase enzymes to conduct single-molecule footprinting, effectively tagging exact nucleotide positions occupied by proteins on the DNA strand. This approach not only pinpoints the precise loci of protein-DNA interactions with near-nucleotide resolution but also preserves the integrity of the DNA sequence, allowing simultaneous genotypic and epigenetic profiling. This dual profiling capability enables researchers to detect how somatic mutations or rare epiallelic variations influence chromatin state and protein occupancy in ways previously elusive to bulk assays.

The implications of being able to study co-binding of proteins along lengthy chromosomal regions in individual cells cannot be overstated. The DAF-seq platform also unlocked the first high-resolution maps revealing cooperative protein occupancy at individual regulatory elements, regions such as promoters, enhancers, and insulators that critically influence transcriptional activity. Indeed, the method demonstrated how proteins cluster or cooperate in situ along the fiber, painting a dynamic picture that informs not just static binding but functional complexes that fine-tune gene expression.

An especially remarkable advancement is the extension of DAF-seq into the single-cell domain, termed single-cell DAF-seq (scDAF-seq). This innovation makes it possible to generate comprehensive chromatin fiber maps spanning 99% of each single cell’s mappable genome. This breadth is unparalleled, offering insights into chromatin states across entire chromosomes rather than limited loci, fundamentally changing the scale at which chromatin architecture can be studied in cellular contexts.

The application of scDAF-seq has exposed a profound level of chromatin plasticity. Research findings indicate that chromatin actuation patterns — the functional occupancy states of proteins along chromatin — diverge by an astonishing 61% between the two haplotypes within a single cell. Even more strikingly, intercellular comparisons reveal a 63% divergence in chromatin actuation among different cells. These revelations underscore the remarkable epigenomic variability that has been suspected but difficult to quantify until the advent of this technology.

Such heterogeneity at the single-fiber level highlights biological processes that could underlie phenomena like allelic imbalance, imprinting, and differential gene expression, which are crucial in development, immune responses, and disease susceptibility. For example, somatic variants that had previously been discounted as irrelevant background noise can now be directly connected to alterations in protein occupancy and regulatory outcomes, deepening our understanding of genotype-phenotype relationships.

One of the most fascinating observations facilitated by scDAF-seq is the preferential co-actuation of regulatory elements along the same chromatin fiber, exhibiting a distance-dependent pattern reminiscent of cohesin-mediated chromatin loops. These loops, long thought to organize chromatin topology and enhance regulatory interactions by bringing distant DNA elements into close proximity, are now revealed to function in concord with protein occupancy states, corroborating models of 3D genome architecture and its impact on transcriptional regulation.

From a methodological perspective, the power of DAF-seq lies in its ability to integrate multiple layers of genetic and epigenetic information on a single DNA molecule. This integration is vital for dissecting complex regulatory networks because chromatin states are often heterogenous and context-dependent. By paralleling nucleotide-level sequence data with chromatin occupancy maps, researchers can now objectively evaluate how variants, both inherited and somatic, impact chromatin dynamics at the molecular scale.

The expansive coverage achieved by DAF-seq—mapping the chromatin fiber architecture over entire chromosomes—provides a broader context for understanding gene regulatory landscapes. Traditional assays have either fallen short in resolution or lacked single-cell granularity, but this technology bridges both gaps. It enables future exploration into how chromatin fiber architecture shifts during cellular differentiation, oncogenesis, or in response to environmental stimuli, offering a platform for novel diagnostic and therapeutic strategies.

Moreover, the potential applications in clinical genomics are substantial. As precision medicine increasingly seeks to understand patient-specific regulatory features influencing disease phenotypes, tools like DAF-seq could provide unmatched insights into somatic mutation impacts and rare epigenetic modifications. This precise molecular detail will be instrumental in deciphering the functional consequences of genetic alterations and how they manifest in cell behavior and pathology.

This research shines a light on the complexity and dynamism of chromatin biology, challenging previous assumptions of uniform chromatin states within diploid cells. By revealing extensive compartmentalization and divergence in regulatory protein binding patterns even within a single cell, DAF-seq invites a re-evaluation of regulatory paradigms and models of chromatin function.

DAF-seq thus marks not merely an incremental improvement but a paradigm shift—a technology capable of revealing the intimate choreography of protein-DNA interactions with unprecedented resolution, scale, and single-cell specificity. This capacity heralds a new era in chromatin biology and genomics, opening pathways to understanding the fundamental mechanics of gene regulation in health and disease with exquisite detail.

In conclusion, the advent of DAF-seq and scDAF-seq introduces a cutting-edge toolkit for scientists probing the epigenomic underpinnings of cellular identity and variability. Moving forward, the integration of such single-molecule and haplotype-aware regulatory maps promises to unravel the intricacies of the genome’s functional organization, heralding transformative insights into cell biology, development, and pathogenesis.

Subject of Research: Chromatin fiber architecture, single-molecule protein occupancy, and single-cell haplotype-resolved gene regulation.

Article Title: Mapping single-cell diploid chromatin fiber architectures using DAF-seq.

Article References:
Swanson, E.G., Mao, Y., Mallory, B.J. et al. Mapping single-cell diploid chromatin fiber architectures using DAF-seq. Nat Biotechnol (2025). https://doi.org/10.1038/s41587-025-02914-3

DOI: https://doi.org/10.1038/s41587-025-02914-3