ATAC-Seq

ATAC-Seq (Assay for Transposase-Accessible Chromatin using Sequencing) is an epigenomic methodology that maps open, accessible euchromatin—including active gene promoters, distal enhancers, silencers, insulators, and nucleosome positions—by probing native nuclear chromatin with an engineered hyperactive Tn5 transposase. Invented in 2013 by Jason D. Buenrostro, Paul G. Giresi, Lisa C. Zaba, Howard Y. Chang, and William J. Greenleaf at Stanford University, ATAC-Seq replaced multi-day chromatin accessibility assays (such as DNase-Seq, MNase-Seq, and FAIRE-Seq) with a two-step enzymatic reaction completed in under three hours.

Within the eukaryotic nucleus, roughly 97% to 99% of the 3.055-gigabase human genome is packaged into compact chromatin fibers, where 147 base pairs of DNA wrap 1.65 times in a left-handed superhelix around octamers of core histone proteins (two copies each of H2A, H2B, H3, and H4). For transcription factors and RNA Polymerase II to initiate transcription, ATP-dependent chromatin remodeling enzymes (such as SWI/SNF and ISWI) must slide or evict nucleosomes to expose nucleosome-free regions (NFRs). By reducing the required sample input by three to four orders of magnitude—from 10⁷–10⁸ cells in classical DNase-Seq down to 500–50,000 bulk cells or individual nuclei in single-cell sequencing (scATAC-seq)—ATAC-Seq enabled routine epigenomic profiling of primary clinical biopsies, sorted immune subsets, and developmental lineages.

Engineered Tn5 Transposome Assembly and Cut-and-Paste Tagmentation

Wild-type prokaryotic Tn5 transposase is a bacterial "cut-and-paste" enzyme that excises transposable elements and inserts them into target DNA at low basal frequency to avoid killing the host bacterium. Building on the biochemical mutagenesis of Igor Goryshin and William Reznikoff (1998)—who introduced three hyperactivity substitutions (E54K, M56A, and L372P) into Tn5—ATAC-Seq pre-loads recombinant hyperactive Tn5 dimers in vitro with synthetic 19-bp Mosaic End (ME) double-stranded oligonucleotides fused to Illumina Read 1 (i5) and Read 2 (i7) sequencing adapters, forming an active transposome.

When isolated, unfixed cell nuclei (prepared in mild non-ionic detergents such as digitonin and Tween-20 under the Omni-ATAC protocol developed by Ryan Corces et al., 2017, which depletes contaminating mitochondrial DNA reads from >50% down to <2%–5%) are incubated with the Tn5 transposome at 37 °C for 30 minutes in a Mg²⁺-containing buffer, steric hindrance governs enzyme access. Because the bulky dimeric Tn5 enzyme (~106 kDa) cannot cleave DNA wrapped around histone octamers or condensed inside heterochromatin, Tn5 inserts selectively into accessible linker DNA and nucleosome-free regions (NFRs). In a single concerted transesterification step termed tagmentation, the 3′-OH termini of the pre-loaded Mosaic End adapters perform nucleophilic attack on both strands of the accessible genomic DNA, simultaneously fragmenting the chromatin and covalently ligating sequencing adapters to the 5′ ends of every cleaved fragment.

Quantitative Comparison of Chromatin Accessibility and Footprinting Assays

The Multimodal Fragment-Length Periodicity Signature

Following tagmentation, purified DNA fragments undergo 5 to 10 cycles of PCR amplification—starting with a 5-minute 72 °C gap-filling extension step where DNA polymerase fills in the 9-bp single-stranded gaps left by Tn5 insertion—and are sequenced via paired-end massively parallel sequencing (30 to 50 million paired reads for peak calling, or >100–200 million reads for transcription factor footprinting).

Because paired-end sequencing measures the genomic distance between the Read 1 and Read 2 Tn5 insertion events, plotting the fragment length distribution of an ATAC-Seq library yields a diagnostic multimodal periodicity ladder:

1. Sub-nucleosomal Peak (<100 bp): Fragments generated when two Tn5 dimers insert inside the same open Nucleosome-Free Region (NFR) at an active promoter or enhancer.
2. Mononucleosome Peak (~180–200 bp): Fragments generated when two Tn5 dimers cleave the accessible linker DNA directly flanking a single intact core nucleosome (147 bp core + linker overhang).
3. Dinucleosome (~380–400 bp) and Trinucleosome (~570–600 bp) Peaks: Fragments spanning two or three adjacent phased nucleosomes.

Correcting the 9-bp Staggered Tn5 Cleavage Offset (+4 bp / −5 bp)

During tagmentation, the dimeric Tn5 transposase binds across the central groove of B-form DNA and cleaves the two antiparallel phosphodiester backbones with a 9-base-pair staggered overhang, centered 4.5 bp inward from the 5′ end of each sequenced read. Consequently, the true center of the Tn5 insertion event is offset from the 5′ alignment coordinate of each read.

In every bioinformatics ATAC-Seq pipeline (ENCODE ATAC-Seq pipeline, MACS2, ArchR, Signac), aligned 5′ read coordinates must be shifted prior to nucleotide-resolution analysis: reads mapping to the positive (+) strand are shifted by +4 bp, and reads mapping to the negative (−) strand are shifted by −5 bp. Quality control of the resulting library is quantified by two ENCODE metrics: the Fraction of Reads in Peaks (FRiP ≥ 0.20–0.30) and the Transcription Start Site (TSS) Enrichment Score (the ratio of sequencing read density centered directly at annotated RefSeq TSSs relative to flanking ±2,000 bp background windows, where mammalian libraries achieve TSS Enrichment ≥ 7 to 15).

Transcription Factor Footprinting and Single-Cell Multi-Omics

Even within an open, nucleosome-free enhancer or promoter peak, a sequence-specific transcription factor (such as CTCF, NRF1, or AP-1) bound tightly to its 8–20 bp DNA recognition motif sterically blocks Tn5 transposase from cleaving the base pairs directly beneath the protein while leaving the flanking open DNA hyper-accessible to tagmentation. When sequenced to high depth (>100 million paired reads), this steric protection creates a narrow nucleotide-level depletion dip ("footprint") inside the broader ATAC-Seq peak. Computational footprinting algorithms—such as TOBIAS (Maren Bentsen et al., 2020) and HINT-ATAC—mathematically deconvolve the intrinsic sequence cleavage bias of Tn5 (using position-weight hexamer bias models) to infer differential transcription factor occupancy across thousands of motifs simultaneously in a single experiment without antibodies (unlike ChIP-Seq).

In single-cell multiome assays (such as 10x Chromium Single Cell Multiome ATAC + Gene Expression or SHARE-seq), isolated nuclei undergo in situ Tn5 tagmentation followed by droplet co-encapsulation that barcodes both tagmented genomic DNA fragments (scATAC-seq) and nuclear polyadenylated mRNAs (snRNA-seq) with the same cell barcode—directly linking distal non-coding enhancer accessibility to target gene transcription within individual cells.

Why did early ATAC-Seq protocols suffer from 50%–80% mitochondrial DNA contamination, and how did Omni-ATAC solve it?

Each mammalian cell contains only two diploid copies of the nuclear genome packaged inside chromatin, but carries hundreds to thousands of copies of the 16.5-kb circular mitochondrial genome (mtDNA), which lacks histones and is therefore 100% open and hyper-accessible to Tn5 tagmentation. In the original 2013 ATAC-Seq lysis protocol (using only IGEPAL CA-630 detergent), outer mitochondrial membranes remained tethered to the endoplasmic reticulum and co-pelleted with nuclei, consuming 50%–80% of sequencing reads. The Omni-ATAC protocol (Corces et al., 2017) combines digitonin (which selectively permeabilizes cholesterol-rich plasma membranes while leaving cholesterol-poor nuclear envelopes intact), Tween-20 (which strips away attached mitochondria), and PBS washes, reducing mtDNA reads to <2%–5%.

Why cannot standard ATAC-Seq be performed on naked purified genomic DNA to identify open chromatin?

Chromatin accessibility is a structural property of intact DNA-histone nucleoprotein assemblies inside the cell nucleus, not of the primary DNA nucleotide sequence itself. Once proteins and histones are stripped away during standard phenol-chloroform or column genomic DNA extraction, the 3-Gb genome becomes uniformly naked and accessible to Tn5 transposase—yielding a flat background library devoid of promoter or enhancer peaks (tagmentation of naked DNA is the basis of commercial Illumina Nextera library prep). ATAC-Seq must therefore be performed on intact, unfixed (or lightly formaldehyde-fixed) nuclei before protein removal.

How does ATAC-Seq differ from ChIP-Seq and Bisulfite Sequencing in clinical epigenomics?

The three core epigenomic sequencing assays interrogate separate layers of gene regulation: bisulfite sequencing maps covalent 5-methylcytosine (5mC) modifications directly on DNA CpG dinucleotides (associated with promoter silencing); ChIP-Seq (and CUT&RUN/CUT&Tag) uses a specific antibody to map the binding locations of one chosen protein or histone mark at a time; whereas ATAC-Seq requires no antibody or prior knowledge of which transcription factors are active, capturing the global profile of all open cis-regulatory elements across the genome in a single 50,000-cell assay and inferring hundreds of active transcription factor networks simultaneously via motif enrichment (chromVAR) and footprinting.

1. Buenrostro, J. D., Giresi, P. G., Zaba, L. C., Chang, H. Y., & Greenleaf, W. J. (2013). "Transposition of Native Chromatin for Fast and Sensitive Epigenomic Profiling of Open Chromatin, DNA-Binding Proteins and Nucleosome Position." Nature Methods, 10(12), 1213–1218. https://doi.org/10.1038/nmeth.2688
2. Corces, M. R., Trevino, A. E., Hamilton, E. G., Greenside, P. G., Sinnott-Armstrong, N. A., Vesuna, S., Satpathy, A. T., Rubin, A. J., Montine, K. S., Wu, B., et al. (2017). "An Improved ATAC-seq Protocol Reduces Background and Enables Interrogation of Frozen Tissues." Nature Methods, 14(10), 959–962. https://doi.org/10.1038/nmeth.4396
3. Buenrostro, J. D., Wu, B., Litzenburger, U. M., Ruff, D., Gonzales, M. L., Snyder, M. P., Chang, H. Y., & Greenleaf, W. J. (2015). "Single-Cell Chromatin Accessibility Reveals Principles of Regulatory Variation." Nature, 523(7561), 486–490. https://doi.org/10.1038/nature14590
4. Bentsen, M., Goymann, P., Schultheis, H., Klee, K., Petrova, A., Wiegandt, R., Fust, A., Preussner, J., Kuenne, C., Braun, T., Kim, J., & Looso, M. (2020). "ATAC-seq Footprinting Unravels Kinetics of Transcription Factor Binding During Zygotic Genome Activation." Nature Communications, 11(1), 4267. https://doi.org/10.1038/s41467-020-18035-1