CUT&Tag (Cleavage Under Targets and Tagmentation) has emerged as one of the most significant cost-saving shifts in chromatin profiling, delivering genome-wide maps of how proteins interact with DNA using a fraction of the starting material and sequencing depth that older methods demand. Where the long-dominant technique, ChIP-seq, typically requires millions of cells and deep sequencing runs to produce usable data, CUT&Tag can generate high-quality profiles from as few as 60 cells, with far less background noise cluttering the results. The method works by guiding a cutting enzyme directly to a protein of interest on intact chromatin, then tagging the surrounding DNA fragments for sequencing in a single streamlined reaction. That simplicity translates into real savings in time, reagents, and sequencing costs, which is why the technique has spread rapidly across labs studying gene regulation in organisms from humans to plants.
Why ChIP-Seq Became a Bottleneck
For years, chromatin immunoprecipitation followed by sequencing (ChIP-seq) was the standard way to figure out where particular proteins or chemical marks sit along the genome. The basic idea is straightforward: chemically glue proteins to DNA, shatter the DNA into small pieces, fish out the fragments attached to your protein of interest with an antibody, then sequence those fragments to see where they came from. The problem is that every step introduces noise. Shattering chromatin randomly creates a huge pool of irrelevant DNA, which means you need to sequence very deeply just to see your signal above the background. Standard protocols call for somewhere between one and twenty million cells per experiment, a requirement that makes many biologically interesting samples simply off-limits.
Think about a rare tumor biopsy, a handful of stem cells from an early embryo, or a small cluster of neurons dissected from a brain slice. These are exactly the samples where understanding chromatin state matters most, and exactly the samples where ChIP-seq struggles. Researchers have developed low-input ChIP-seq workarounds over the years, but these tend to be finicky, expensive, and still limited compared to what CUT&Tag now offers.
How the Method Actually Works
CUT&Tag skips the brute-force approach of ChIP-seq entirely. Instead of fragmenting the entire genome and hoping to pull out relevant pieces afterward, it targets its cutting directly to the sites where a protein of interest sits. The workflow starts with intact cells or nuclei bound loosely to magnetic beads. You add an antibody that recognizes your target, whether that is a histone modification like H3K27me3 or a transcription factor like CTCF. Then you add a fusion protein: a Protein A (or Protein G) domain fused to a hyperactive Tn5 transposase that has been pre-loaded with sequencing adapters.
The Protein A half of this fusion binds to the antibody, tethering the transposase right next to the chromatin feature you care about. When you activate the transposase with magnesium, it cuts the nearby DNA and simultaneously inserts sequencing adapters into the cut ends. This “tagmentation” step means the fragments are essentially ready for PCR amplification and sequencing the moment they are released. There is no sonication, no lengthy immunoprecipitation, and no adapter ligation step. The whole bench protocol can be completed in a single day.
Because the enzyme only cuts near tethered sites, the vast majority of sequenced fragments carry real signal rather than noise. Direct comparisons show that CUT&Tag profiles accumulate signal at known chromatin features far more efficiently than ChIP-seq, which means you can get away with sequencing far fewer reads and still see clear patterns.
Signal Quality and Sequencing Savings
The practical payoff of targeted cutting is dramatic. When researchers compared CUT&Tag, CUT&RUN (a related method using a different enzyme), and ChIP-seq side by side for the same chromatin marks, background noise dominated the ChIP-seq data, meaning much of the sequencing budget was spent reading uninformative fragments. Both CUT&RUN and CUT&Tag had extremely low background noise, but CUT&Tag showed the greatest signal accumulation at known binding sites, implying it would be most effective at distinguishing chromatin features with the fewest reads.
In concrete terms, this means a CUT&Tag experiment might need only a few million sequencing reads to produce a publication-quality genome-wide map, whereas a ChIP-seq experiment targeting the same mark could require tens of millions of reads to achieve comparable clarity. Since sequencing costs are often the largest line item in a chromatin profiling experiment, this difference alone can cut per-sample costs by an order of magnitude. For labs running dozens or hundreds of samples in a study, those savings are transformative.
Working With Tiny Amounts of Material
Perhaps the most striking advantage of CUT&Tag is how little starting material it needs. In the original benchmarking work, researchers profiled the repressive histone mark H3K27me3 across a roughly 1,500-fold range of input, from 100,000 cells all the way down to just 60 cells. The chromatin profiles remained very similar and high quality across that entire range.
This opens the door to experiments that were previously impractical. Clinical biopsies that yield only a few thousand cells, sorted immune cell subsets, rare developmental progenitors, even individual embryos can now be profiled for their chromatin landscape without pooling samples or amplifying material through multiple rounds of processing. In one demonstration using bovine blastocysts, genome-wide maps of both H3K4me3 (an activating mark) and H3K27me3 (a repressive mark) were generated from single embryos, producing profiles that broadly agreed with conventional ChIP-seq data.
Histone Marks Versus Transcription Factors
CUT&Tag works best, and most reliably, for histone modifications. These marks are abundant: every nucleosome carries histone tails that can be chemically modified, so there are millions of potential target sites spread across the genome. Antibodies against common histone marks like H3K27me3, H3K4me3, and H3K36me2 tend to produce clean, reproducible CUT&Tag profiles with relatively little optimization.
Transcription factors are a different story. These proteins are far less abundant than histone marks, and many of them bind DNA transiently or weakly. Because CUT&Tag relies on the antibody-transposase complex finding and staying put at its target, proteins that hop on and off DNA quickly can be hard to capture. Current CUT&Tag methods for profiling specific transcription factor-DNA interactions remain technically challenging precisely because of this low abundance issue.
One workaround is light chemical crosslinking to stabilize protein-DNA contacts before running the protocol. Mild formaldehyde treatment has been used for this purpose but does not always improve results and can actually reduce library yields at higher concentrations. More recently, a protein-protein crosslinker called EGS has shown promise for stabilizing interactions without the drawbacks of formaldehyde, owing to its longer chemical spacer arm that does not crosslink DNA as aggressively. Optimizing crosslinking conditions for each transcription factor remains an active area of protocol development.
Single-Cell Chromatin Profiling
The low material requirements of CUT&Tag naturally lend themselves to single-cell applications, where each “sample” is one cell’s worth of chromatin. Several groups have adapted the chemistry for droplet-based or combinatorial-indexing single-cell workflows. The appeal is obvious: instead of seeing the average chromatin state across thousands of cells, you can map how chromatin differs from one cell to the next within a tissue, revealing the epigenetic heterogeneity that underlies cell identity and disease.
The challenge is sparsity. A single cell has only two copies of each chromosome, so even a perfectly efficient experiment captures only a thin slice of the possible signal. Integrating data across many cells and multiple histone marks requires sophisticated computational methods. One approach, called scCUT&Tag-pro, addresses this by simultaneously measuring chromatin modifications and cell-surface proteins from the same single cell, giving researchers two independent layers of information to help identify cell types and states.
Combining histone modification data from multiple single-cell experiments into unified chromatin state maps is another frontier. Computational frameworks have been developed that infer and annotate chromatin states based on combinatorial histone modification patterns across single cells, effectively recreating at single-cell resolution the kind of chromatin state models that were previously possible only from bulk sequencing of millions of cells.
A Known Artifact Worth Understanding
No method is perfect, and CUT&Tag has a well-characterized bias that users need to know about. Because the Tn5 transposase has an inherent preference for cutting open, accessible chromatin, it can produce spurious signal at actively transcribed gene promoters even when the antibody targets a repressive mark like H3K27me3. In one systematic analysis, roughly 18% of CUT&Tag peaks for H3K27me3 did not overlap with ChIP-seq peaks for the same mark. When researchers examined these CUT&Tag-unique signals, they found clear enrichment near the promoters of actively transcribed genes, exactly the kind of place where a repressive mark should not appear.
This open-chromatin bias does not invalidate CUT&Tag results, but it does mean that raw peak calls need to be interpreted carefully, especially in regions of high transcriptional activity. Computational correction tools like PATTY have been developed specifically to account for this bias in both bulk and single-cell CUT&Tag data. For labs moving from ChIP-seq to CUT&Tag, awareness of this artifact and appropriate bioinformatic filtering are important parts of getting reliable results.
Bioinformatic Tools Tailored to CUT&Tag Data
Standard ChIP-seq analysis pipelines were designed around the assumption of high background noise and deeply sequenced libraries. Applying those same tools to CUT&Tag data, which has a very different signal-to-noise profile, can produce misleading results. Peak callers designed for ChIP-seq sometimes struggle with the sharp, low-background peaks that CUT&Tag produces, or they may not handle the broad diffuse domains characteristic of marks like H3K27me3 as well as purpose-built alternatives.
SEACR (Sparse Enrichment Analysis for CUT&RUN) was one of the first peak callers designed for the high signal-to-noise data produced by antibody-directed cleavage methods. It uses a data-driven threshold rather than requiring a predefined background model, which suits the sparse background of CUT&Tag well. GoPeaks is another tool built specifically for histone modification CUT&Tag data, designed to handle the range of peak profiles produced by different chromatin marks, from the tight peaks of active promoter marks to the broad domains of repressive modifications. Choosing the right peak caller for your specific mark and experimental design is not a minor detail; it can substantially affect which genomic regions are called as enriched and which are missed.
Adapting the Protocol for Plant Biology
CUT&Tag was developed and first validated in mammalian cells, but it has been adapted with notable success for plant systems. Plants pose particular challenges for chromatin profiling: tough cell walls must be disrupted to release nuclei, and many plant species lack the well-characterized antibody reagents available for human or mouse research. Despite these hurdles, CUT&Tag has been validated in the model plant Arabidopsis as a reliable and cost-effective method for epigenomic profiling, compatible with limited amounts of tissue and offering higher resolution than ChIP-seq.
Because the CUT&Tag protocol starts from isolated nuclei rather than whole cells, it can in principle be adapted to any organism from which clean nuclei can be prepared. This makes it applicable to both model plants with well-annotated genomes and non-model species where genomic resources are still being developed. For plant biologists studying crop epigenetics or environmental stress responses, this accessibility is a meaningful advance.
Reagent flexibility also matters in non-mammalian systems. The original CUT&Tag fusion protein uses Protein A, which binds well to rabbit antibodies but poorly to mouse IgG1 antibodies commonly used in plant research. Protein G-Tn5 fusion proteins have been engineered as alternatives, and direct comparisons in Arabidopsis showed that the Protein G version produced better-quality libraries when paired with mouse IgG1 antibodies. Having both Protein A and Protein G versions available lets researchers choose the fusion protein that best matches their antibody species, a practical detail that makes the method more broadly useful across experimental systems.
Spatial Chromatin Profiling in Intact Tissue
One of the more ambitious extensions of CUT&Tag moves beyond dissociated cells entirely. Spatial-CUT&Tag combines the in situ tagmentation chemistry with microfluidic barcoding to map histone modifications across tissue sections while preserving spatial information. Instead of losing track of where each cell sat in the original tissue, this approach encodes positional coordinates into the sequencing data.
In demonstrations using mouse embryo tissue sections, spatial-CUT&Tag revealed tissue-type-specific chromatin states that matched reference datasets, and provided spatial patterning of cell types determined by histone modifications in the mouse brain, including epigenetic control of cortical layer development. By identifying 20-micrometer pixels containing only one nucleus through immunofluorescence imaging, single-cell-resolution epigenomic data could be extracted in situ.
Spatial chromatin profiling is still in its early stages as a technology, and throughput and resolution remain limited compared to single-cell RNA sequencing methods. But the ability to see not just which chromatin states exist in a tissue but where they exist adds a dimension of information that bulk or dissociated single-cell methods fundamentally cannot provide. For developmental biology and tumor microenvironment studies, where the spatial arrangement of cell types carries biological meaning, this is a qualitatively different kind of data.
Practical Considerations for Getting Started
For a lab considering CUT&Tag for the first time, a few practical realities are worth knowing. The reagents are relatively inexpensive compared to ChIP-seq: there is no need for sonication equipment, DNA size-selection columns are often unnecessary because tagmentation produces fragments in a useful size range, and the reduced sequencing depth requirement shrinks the single largest cost. Many groups prepare their own Protein A-Tn5 or Protein G-Tn5 fusion proteins in-house from publicly available expression constructs, which further reduces per-experiment costs compared to purchasing commercial kits.
Antibody quality is, if anything, more important for CUT&Tag than for ChIP-seq. Because the transposase cuts wherever the antibody tethers it, a non-specific antibody will generate non-specific cuts that look like real signal in the low-background CUT&Tag data. In ChIP-seq, non-specific pull-down fragments tend to drown in the general noise; in CUT&Tag, they can masquerade as genuine peaks. Validating antibody specificity and using well-characterized, lot-tested antibodies is not optional.
The protocol also benefits from careful attention to cell or nucleus handling. Concanavalin A-coated magnetic beads are used to immobilize cells or nuclei, and inconsistent bead binding can introduce variability between replicates. Gentle washing steps preserve chromatin integrity, since harsh detergent conditions or prolonged incubations can degrade the very structures you are trying to map. The bench protocol is genuinely simpler than ChIP-seq, but “simpler” does not mean “forgiving.” Small deviations in timing, temperature, or reagent concentrations can meaningfully affect data quality, especially for difficult targets like transcription factors.
Where Systematic Benchmarking Stands
As CUT&Tag has matured from a novel technique to a mainstream tool, systematic benchmarking efforts have become increasingly important. Recent large-scale comparisons have evaluated how different protocol variations, such as crosslinking conditions, antibody concentrations, and transposase preparations, affect the reliability and reproducibility of the resulting chromatin maps. These studies highlight that seemingly minor protocol choices can shift which peaks are detected and how well replicates agree with one another.
The crosslinking question is a good example of how benchmarking changes practice. For histone marks, native (no crosslinking) CUT&Tag generally works well. For transcription factors, light crosslinking improves detection, but the optimal crosslinker and concentration depend on the specific factor being profiled. EGS crosslinking has been shown to improve discovery of CTCF-binding sites in systematic tests, leading to protocol updates that now recommend it for certain targets. Without careful benchmarking, labs might assume a single protocol works for all targets and miss real binding events or call false ones.
The field is still settling on best practices, and protocol recommendations continue to evolve. For anyone reading a CUT&Tag paper, it is worth checking the methods section for details on crosslinking, antibody lot, Tn5 preparation, and peak-calling software, since all of these can influence what the data show. The technique is robust and reproducible when executed well, but “executed well” means paying attention to the specifics rather than treating the protocol as a one-size-fits-all recipe.