Autonomous Replication in eccDNA: Evidence, Open Questions, and How Sequencing Can Help

Are extrachromosomal circular DNAs (eccDNAs) just passive debris from genome instability—or do some of them autonomously replicate and persist as bona fide episomes? This guide takes a clear, argued position: there is strong, emerging evidence that at least a subset of eccDNAs can support replication-like behaviors, and the field now has the tools to obtain decisive proof in human cells. We also spell out the gaps and the exact experiments that can close them.

Throughout, we distinguish small eccDNA (often called microDNA; generally <10 kb) from large cancer-associated circular amplicons (ecDNA; tens of kilobases to megabases). Our focus is small eccDNAs and the question of autonomous replication in eccDNA as a mechanism for persistence and functional impact. We anchor the argument in replication biology, cross-system demonstrations of origin activity, and sequencing-led workflows that can verify or falsify autonomous replication with orthogonal evidence.

What counts as autonomous replication in eccDNA?

By autonomous replication in eccDNA, we mean post-formation maintenance where a circular DNA molecule initiates DNA replication at its own origin-like sequence and supports fork progression independently of its chromosomal locus. This is distinct from replication-dependent biogenesis (for example, when replication stress helps create eccDNA via break-induced replication) and from rolling-circle amplification in vitro. Evidence for autonomous replication should demonstrate:

• Origin firing on the circle (or origin-factor recruitment to circle sequences) under normal cell-cycle conditions.
• Replication intermediates characteristic of circular templates (bubble arcs/theta structures) in cellular DNA.
• Incorporation of replication labels (BrdU/EdU) into the circular fraction, above matched background.
• Orthogonal sequencing confirmation that the labeled or structured molecules are bona fide circles with defined junctions.

These criteria collectively go beyond "circles exist" to "circles replicate." They are intentionally conservative to avoid mistaking biogenesis correlates for maintenance.

The evidence we have so far (human cells and cross-system clues)

A growing set of observations supports the plausibility of autonomous replication in eccDNA, even if small-circle, human-cell, peer-reviewed proofs remain limited.

Human-cell biogenesis is replication-associated. In 2024, Gadgil and colleagues showed that ectopic, non-B DNA–forming microsatellites integrated in human cells generate eccDNAs via replication-dependent mechanisms consistent with break-induced replication and template switching. Mutagenesis spread for several kilobases, and replication stress agents shifted eccDNA profiles. These findings make it hard to argue that replication is irrelevant to eccDNA biology, even if they don't yet prove that the circles, once formed, keep replicating on their own. See the detailed demonstration in the NAR Cancer study by Gadgil et al. (2024), which provides strong replication-linked biogenesis evidence and a methodological foundation for origin-focused assays in human systems.

Cross-system demonstrations of functional origins on eccDNA-like replicons. Outside human small eccDNA, there's direct functional evidence that circle-borne sequences can act as origins. In plants, origin-like segments from the Amaranthus palmeri eccDNA replicon enabled autonomous replication in yeast when cloned into ARS-less vectors. That study by Molin et al. (2020) validated that some extrachromosomal replicons carry bona fide autonomously replicating sequences (ARS) with expected sequence features (AT-richness, DNA unwinding elements). In yeast itself, a substantial fraction of eccDNAs harbor ARS/ACS motifs and can replicate and propagate under selection. Together, these data prove the principle: if an eccDNA contains an origin that is competent in a eukaryotic replication system, autonomous replication is possible.

Human cancer ecDNA as a conceptual comparator. While our focus is small eccDNA, large ecDNA in human cancer cells clearly engages the replication machinery. Studies of ecDNA show redistributed origins and vulnerable forks under stress, underscoring that circular DNA can be a substrate for replication initiation and fork movement in human cells. The topology and size are different, but the mechanistic message is relevant: circular templates are not invisible to the replication program.

Where "eccDNA Michael Leffak" fits. Michael Leffak and colleagues have long advanced the study of replication origins in mammalian chromosomes, especially the sequence and chromatin features that favor initiation. While we did not find peer-reviewed, human-cell papers that explicitly demonstrate "eccdna michael leffak replication" for small circles, that body of origin biology informs exactly how to test eccDNAs for initiation capacity: enrich for AT-rich/DUE sequences, map ORC/MCM binding, and measure nascent strand signal. In other words, the replication-origin framework attributed to Leffak's field offers the logic and assays to prosecute the autonomous replication question in eccDNA without overclaiming.

The bottom line on evidence. The combination of (1) replication-dependent biogenesis in human cells, (2) functional origin activity on eccDNA replicons across systems, and (3) clear replication of large circular DNA in human cancer cells adds up to strong, convergent support that some small eccDNAs could autonomously replicate. What's missing is a decisive, multi-assay demonstration in human cells that ties origin occupancy, replication labeling, and circular structure to the same molecules. That's what the next sections enable you to design.

Methods to detect replication on eccDNAs (from hypothesis to proof)

To move from plausible to proven, you'll need orthogonal assays that converge on the same set of circles. Think of it as a three-legged stool: biochemical structure, replication labeling, and sequencing confirmation.

BrdU/EdU pulldown integrated with Circle-Seq

Concept. BrdU (or EdU) pulse-label during S phase and immunoprecipitate nascent DNA, then enrich for circles and sequence them. If circle-borne DNA is enriched in the labeled fraction relative to input, that's direct evidence that the circle sequences were copied during the pulse window.

Suggested design and controls (RUO):

• Cell-cycle control. Synchronize cells to narrow the S-phase window (e.g., double-thymidine block or thymidine–nocodazole). Use short BrdU pulses (15–30 minutes) to enrich for initiation-proximal nascent DNA.
• Immunoprecipitation. Extract genomic DNA under gentle conditions, denature as needed for BrdU-IP, and immunoprecipitate labeled strands with validated anti-BrdU antibodies. Perform an EdU-click variant if desired; both are acceptable with adequate optimization. For context on nascent-strand IP strategies, see studies mapping replication initiation and BrdU-labeled DNA on chromosomal templates, such as the nascent-strand and BrdU-IP literature in Nature Communications (Liu et al., 2021) and related resources.
• Circle enrichment post-IP. Treat IP'd DNA with exonucleases that degrade linear DNA (e.g., Plasmid-Safe DNase or ExoV conditions) to spare circular templates. Then perform rolling-circle amplification (RCA; phi29) to increase yield. RCA adds bias; measure and report it.
• Circle-Seq library. Prepare Circle-Seq libraries from the IP'd, circle-enriched fraction and from matched input. Sequence to sufficient depth for robust junction detection. Validate circle junctions by long-read reads where possible (more below).
• Controls. Include a spike-in plasmid with a known, firing origin as a positive control for BrdU incorporation and 2D-gel validation; include an origin-defective plasmid or nicked circle as a negative control. Add mitochondrial DNA contamination checks and a linear DNA spike-in to confirm exonuclease selectivity.
• Analysis. Compare IP/input enrichment for junction-spanning reads assigned to specific circles. A 2–3× enrichment at circle-junction qPCR or junction read counts is a reasonable starting threshold, to be refined by your lab.

Neutral protocols. For a standardized eccDNA purification workflow, follow JoVE's 2016 "Genome-wide Purification of Extrachromosomal Circular DNA" Circle-Seq protocol, which details exonuclease digestion, rolling-circle amplification, and library preparation. For replication labeling and origin-proximal DNA capture, see Quantitative Bromodeoxyuridine Immunoprecipitation Analyzed by High‑Throughput Sequencing (qBrdU‑Seq / QBU), Methods in Molecular Biology (2018)and Methods in Molecular Biology's 2014 protocol for isolation of nascent DNA strands, which can be adapted to assess initiation-capable sequences on circular templates.

Study design tie-in. Drug treatments that modulate replication fork dynamics can clarify signal origins and stress-responsiveness. For study design considerations around hydroxyurea, S-phase timing, and fork stalling, see Replication Stress and eccDNA: Hydroxyurea, Cell Cycle Effects, and Study Design Considerations.

Two-dimensional (2D) neutral gels and Southern confirmation

Concept. 2D neutral agarose gels resolve replication intermediates by size and shape, revealing diagnostic arcs for bubble structures (origin firing) and Y-shaped forks. If you probe a circle-specific junction fragment and detect bubble arcs in the eccDNA-enriched fraction, you've captured replication intermediates from circular templates.

Key parameters and controls (RUO):

• Template preparation. Purify circular DNA-enriched fractions using exonuclease digestion of linear DNA. Validate enrichment by qPCR for known circular junctions versus linear loci.
• Restriction mapping. Choose restriction enzymes that cut the circle once (for Y arcs) or not at all (for bubble arcs), depending on your design. Use plasmid spike-ins with known origins as run controls.
• Gel conditions. Run first dimension at low agarose and low voltage, second dimension at higher agarose and voltage with ethidium bromide to sharpen shape-based separation. Transfer and probe by Southern with oligos spanning circle junctions or internal circle sequences.
• Interpretation. A bubble arc indicates initiation within the fragment; a Y arc indicates replication traversing the fragment from an external origin. On small circles, theta-like intermediates can appear as bubble arcs when the entire circle migrates as a single fragment.

Below is an educational schematic of the expected arcs. In your experiments, replace schematics with actual blots and include plasmid controls for calibration.

Figure 1.2D gel electrophoresis resolves replication intermediates. The bubble arc (highlighted) is a classic signature of initiation within the probed fragment. Schematic for educational purposes.

Origin mapping via ORC/MCM ChIP-seq or nascent-strand sequencing

Concept. If eccDNA coordinates overlap genomic regions that recruit origin recognition complex (ORC), mini-chromosome maintenance (MCM) helicase, or nascent strands, that strengthens the case for initiation-capable sequences on the circle. While we did not identify peer-reviewed human small-eccDNA datasets demonstrating such overlap, the analysis is feasible with public data.

Analysis plan (RUO):

• Collect origin maps. Assemble ORC/MCM ChIP-seq peaks and nascent-strand maps for your cell type or the nearest available proxy from public resources such as the ENCODE portal. See Urban et al. (2015) for a review of origin mapping strategies and features in multicellular eukaryotes.
• Map circle coordinates. Call eccDNA coordinates from Circle-Seq (or long-read assemblies) and represent them as intervals. For circles derived from repeats, use cautious multi-mapping handling.
• Overlap and enrichment. Compute enrichment of circle intervals within ORC/MCM/nascent-strand peaks against size-matched, GC-matched background intervals. Apply multiple-hypothesis correction and report effect sizes and confidence intervals.

Practical ENCODE navigation (RUO). For reproducible public data sourcing, start at the ENCODE Portal's filtered collections: use the ENCODE portal — search for ChIP‑seq experiments (use the left‑hand filters to select replication factors such as ORC/MCM)to locate ORC/MCM datasets, and the ENCODE Repli‑seq/OK‑seq nascent‑strand timing collection for replication‑timing/nascent‑strand tracks (ENCODE Portal, 2024–2026). Filter by organism, biosample (cell line/tissue), and assay to match your system, then download processed bigWig/bed files and metadata (lab, replicate, antibody, audit status) to ensure comparability with your eccDNA intervals.

To illustrate the concept, here's a stylized browser view that overlays eccDNA intervals on ORC/MCM tracks.

Figure 2.Illustrative ORC/MCM occupancy with eccDNA intervals (green boxes). In practice, use matched cell-line datasets and rigorous background models. Figure is illustrative.

A theta replication fork schematic for circles

For clarity, the following self-drawn diagram shows a bidirectional theta replication event on a circular template—a structure that 2D gels can capture as a bubble arc when probed appropriately.

Figure 3.Theta replication on an eccDNA. Ori indicates initiation; leading and lagging strands are labeled; fork arrows show progression. Schematic for educational purposes.

Sequencing and bioinformatics blueprint for replication verification

Sequencing is the arbiter that ties structure and labeling back to specific circles. Here's a practical blueprint to go from raw data to defensible claims—one that integrates short- and long-read evidence and guards against artifacts.

Library strategies (RUO):

• Circle-Seq for discovery. Use exonuclease digestion to remove linear DNA and construct Circle-Seq libraries for both input and BrdU/EdU-IP fractions. Short-read depth should be sufficient to robustly identify junction-spanning reads and quantify enrichment (e.g., 100–200 million read pairs per condition, adjusted for sample complexity).
• Long-read confirmation. Add Oxford Nanopore or PacBio HiFi to directly span circle junctions and assess size heterogeneity. Long reads are invaluable for confirming that your BrdU-enriched molecules are truly circular and not concatemeric artifacts.
• Optional WGS context. Whole-genome sequencing of the matched bulk sample can provide background copy-number context and aid in mapping eccDNA origins when circles are derived from complex loci. For platform-agnostic sequencing design considerations, see CD Genomics' Whole Genome Sequencing page for workflow and deliverable guidance.

Disclosure: CD Genomics is the provider of the sequencing and bioinformatics services referenced here. All methods and service descriptions are offered for research use only (RUO) and are presented as methodological examples, not commercial endorsements or clinical advice; investigators should validate protocols locally before use.

Bioinformatics pipeline (RUO):

• Pre-processing and mapping. Trim adapters; map reads with sensitive settings; retain split reads and discordant pairs. For Circle-Seq, use tools suited to circle detection and junction calling (e.g., ECCsplorer for small eccDNAs; AmpliconArchitect for larger circular amplicons/ecDNA contexts).
• Junction calling and validation. Require multiple unique read supports per junction; corroborate with long-read molecules that cleanly traverse the junction. Quantify junction coverage and report confidence.
• Artifact filters. Remove likely PCR chimeras (read-pair geometry and duplication signatures), tandem-duplication false positives, and mitochondrial DNA contamination. Explicitly document each filter's rationale.
• Enrichment analysis. Compute IP/input enrichment at circle junctions (read-level) and at circle intervals (coverage-level). Report fold-change distributions and confidence intervals.
• Origin overlap statistics. Overlay circle intervals with ORC/MCM/nascent-strand peaks and report enrichment against matched backgrounds.

Pipeline setup and reporting. Reproducible, publication-ready analysis standards help reviewers and collaborators evaluate your claims. For a detailed discussion of detection algorithms, artifact filtering, and reporting standards tailored to eccDNA, consider engaging a specialist team; see CD Genomics' Bioinformatics Services for example pipeline capabilities and RUO deliverables.

Biological consequences if eccDNAs replicate autonomously

If some eccDNAs do replicate autonomously in human cells, several consequences follow.

Persistence of amplified functions. Circles carrying beneficial genes or regulatory elements could persist even if the chromosomal source is lost. This is clearly true for large ecDNA with oncogenes; for small eccDNA, drug-resistance cassettes or promoters/enhancers could, in principle, be maintained on replicating circles. Studies of resistance dynamics in other systems and the plant replicon proof-of-principle suggest how such maintenance could operate. For a broader discussion of eccDNA's role in tumor models and amplification, see the deeper application guide: eccDNA in Cancer: Gene Amplification, Oncogene Regulation, and Research Applications.

Adaptive dynamics under stress. Replication stress modulates circle formation, fork stability, and mutagenesis. Autonomous replication in eccDNA would add another layer: circles could copy (or fail to copy) under specific stress regimes, changing their abundance. Experimental designs using hydroxyurea or polymerase inhibitors should therefore include time-resolved sampling and matched controls.

Mechanistic connections to instability. Autonomous replication requires initiation-competent sequences and the recruitment of origin machinery. That ties directly to formation pathways involving alt-EJ, replication fork collapse, and retrotransposon-associated structures. For a mechanism-forward perspective that sets up origin testing on circles, see Linking eccDNA to Genome Instability: alt-EJ, Replication Stress, and Retrotransposons.

Interpreting "eccDNA Michael Leffak replication" models. Work in the mammalian origin field (often associated with Michael Leffak's contributions to human replication origin mapping and sequence determinants) emphasizes AT-rich DNA unwinding elements, chromatin openness, and transcription–replication crosstalk as predictors of initiation. Applying those principles to eccDNA suggests testable hypotheses: circles enriched for DUE-like segments and favorable chromatin marks should show higher BrdU-IP enrichment and ORC/MCM overlap. The phrase "eccDNA Michael Leffak" here is used as a shorthand for origin-centric replication frameworks that guide practical assay design.

Practical example (RUO): designing a replication-positive eccDNA study

A worked example clarifies design decisions and deliverables. The scenario below is illustrative and keeps claims modest while meeting rigorous criteria.

Study question. Do small eccDNAs in a human cell line carry initiation-competent sequences and replicate autonomously during S phase? In other words, can we obtain direct evidence consistent with "eccDNA Michael Leffak replication" expectations in human cells?

Overview of workflow (RUO):

• Sample and synchronization. Choose a well-characterized human cell line with public origin maps. Synchronize to enrich early S phase.
• BrdU pulse and IP. Pulse BrdU for 20 minutes. Extract DNA and perform BrdU-IP under conditions validated on plasmid controls.
• Circle enrichment. Treat IP'd DNA with Plasmid-Safe DNase, validate retention of plasmid spike-in, and quantify circular enrichment by qPCR.
• Circle-Seq and WGS. Prepare Circle-Seq libraries from IP and input. Optionally sequence matched WGS for context on copy-number and repeats. For sequencing design and standard RUO deliverables (FASTQ/BAM/VCF, QC reports), see CD Genomics' Whole Genome Sequencing page.
• Long-read junction confirmation. Sequence the circle-enriched fractions with ONT or PacBio HiFi to span junctions and verify circle sizes.
• 2D gels and Southern. In parallel, run 2D neutral gels on circle-enriched fractions and probe for several high-confidence circle junctions. Include plasmid controls with known origins to calibrate bubble/Y arc expectations.
• Origin overlap analysis. Overlay circle intervals with ORC/MCM peaks from a matched dataset (or nearest proxy). Compute enrichment with matched backgrounds.
• Reporting and QC. Pre-register analysis thresholds: BrdU-IP enrichment at junctions ≥3× over input; 2D bubble arc detection for at least one junction-probed circle; long-read confirmation of the same circles; significant ORC/MCM overlap (e.g., FDR < 0.05; enrichment Z > 2). Document artifact filters and negative controls. For standardized QC metrics and reproducibility benchmarks, consult the Quality Metrics guidance: Quality Metrics for eccDNA Sequencing.

Where neutral outside support can help (RUO). Teams often divide responsibilities: wet-lab execution in-house, and external support for high-throughput sequencing and pipeline implementation. Under a RUO framework, you can request end-to-end documentation of parameters, software versions, and QC from a provider's bioinformatics unit. See CD Genomics' Bioinformatics Services for typical deliverables, and consult the series article on QC for field-tested metrics.

Criteria for calling "replication-positive."

• Orthogonal convergence. At least one circle shows (a) 2D bubble arc when probed at its junction or internal fragment, (b) ≥3× BrdU-IP enrichment at the same circle's junction relative to input, (c) long-read molecules spanning the circle junction, and (d) significant overlap with origin-factor binding or nascent-strand signals in matched or proxy datasets.
• Reproducibility. The same circle recurs across biological replicates and passages, and QC shows stable enrichment and low artifact rates.
• Specificity. Negative controls (origin-defective plasmid, linear DNA spike-ins) do not exhibit bubble arcs or BrdU-IP enrichment after circle enrichment.

Open questions and priority experiments

To definitively resolve autonomous replication in eccDNA, prioritize experiments that combine assays in the same biological samples and predefine decision thresholds.

• Time-resolved BrdU/EdU-IP + Circle-Seq with synchronized windows to capture initiation-proximal labeling on circles; include matched 2D gels for the same samples.
• ORC/MCM ChIP-seq and nascent-strand mapping in the same cell line used for Circle-Seq, enabling direct origin overlap tests for circle intervals.
• Sequence-feature analysis of circle cohorts that do vs. don't show BrdU enrichment, testing hypotheses inspired by mammalian origin biology (AT content, DUEs, G4 motifs, transcriptional context) associated with the "eccDNA Michael Leffak replication" framing.
• Replication stress titration (e.g., hydroxyurea dose-response) to map how fork dynamics modulate circle maintenance, guided by the replication-stress study-design notes linked above.
• Single-molecule labeling (e.g., ONT BrdU-detection protocols) on circle-enriched fractions to visualize nascent-strand incorporation directly on individual circles.

Conclusion: why sequencing can deliver definitive answers

The case for autonomous replication in eccDNA is getting stronger: replication-dependent biogenesis in human cells, cross-system demonstrations of circle-borne origins, and clear replication of large human circular amplicons together point in the same direction. What the community needs now is convergence—BrdU/EdU labeling, 2D-gel replication intermediates, origin-factor overlap, and junction-resolved sequencing—on the same circles in the same samples.

That's achievable with today's methods. Start with cautious definitions, add rigorous controls, and insist on orthogonal validation. Use Circle-Seq plus long-read confirmation to tie labels and structures back to specific circles; quantify enrichment and overlap with transparent statistics; and publish with complete QC and reporting standards. For pipeline configuration, artifact filtering, and publication-ready figures under RUO standards, you can engage a neutral provider's analysis team. Keep every claim tied to data, and you'll move the field from plausible to proven.

Author

Yang H. — Senior Scientist, CD Genomics; University of Florida.

Yang is a genomics researcher with over 10 years of research experience in genetics, molecular and cellular biology, sequencing workflows, and bioinformatic analysis. Skilled in both laboratory techniques and data interpretation, Yang supports RUO study design and NGS-based projects.

References (with DOIs)

1. Gadgil RY, Kim S, Raghuram N, et al. Microsatellite break-induced replication generates highly mutagenized extrachromosomal circular DNAs. NAR Cancer. 2024;6(2):zcae027. DOI: 10.1093/narcan/zcae027. https://academic.oup.com/narcancer/article/6/2/zcae027/7689927
2. Molin WT, Yaguchi A, Blenner M, Saski CA. Autonomous replication sequences from the Amaranthus palmeri eccDNA replicon enable replication in yeast. BMC Res Notes. 2020;13:330. DOI: 10.1186/s13104-020-05169-0. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7350638/
3. Møller HD, Parsons L, Jørgensen TS, et al. Extrachromosomal circular DNA is common in yeast. Proc Natl Acad Sci USA. 2015;112(24):E3119–E3127. DOI: 10.1073/pnas.1508825112. https://www.pnas.org/doi/10.1073/pnas.1508825112
4. Zhao Y, Chen J, Li G, et al. Extrachromosomal circular DNA: Current status and future perspectives. eLife. 2022;11:e81412. DOI: 10.7554/eLife.81412. https://elifesciences.org/articles/81412
5. Yang L, Liu Q, Zhang X, et al. Extrachromosomal circular DNA: biogenesis, structure, functions and diseases. Signal Transduct Target Ther. 2022;7:211. DOI: 10.1038/s41392-022-01176-8. https://www.nature.com/articles/s41392-022-01176-8
6. Urban JM, Foulk MS, Casella C, Gerbi SA. The hunt for origins of DNA replication in multicellular eukaryotes. Chromosome Res. 2015;23(1):1–26. DOI: 10.1007/s10577-015-9476-4. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4371235/
7. Liu Y, Zhou J, White J, et al. Transcription shapes DNA replication initiation to preserve genome stability. Nat Commun. 2021;12:2113. DOI: 10.1038/s41467-021-23687-3. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8188667/
8. Cebrián J, Castán A, Martínez-Robles ML, et al. Two-Dimensional Gel Electrophoresis to Study the Activity of DNA Topoisomerases in vitro. Int J Mol Sci. 2021;22(22):12369. DOI: 10.3390/ijms222212369. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8615216/
9. Hashimoto Y, Puddu F, Costanzo V. ATM and ATR Activities Maintain Replication Fork Integrity during SV40 Chromatin Replication under Stress. PLoS Pathog. 2013;9(4):e1003283. DOI: 10.1371/journal.ppat.1003283. https://journals.plos.org/plospathogens/article?id=10.1371/journal.ppat.1003283
10. Wang G, Chen X, Zhang X, et al. Methods to detect replication-dependent DNA structures via two-dimensional gel electrophoresis. Methods. 2013;67(2):242–248. DOI: 10.1016/j.ymeth.2013.05.001. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3842213/
11. Saini N, Ramakrishnan S, Elango R, et al. Migrating bubble during break-induced replication. Nature. 2013;502(7471):389–392. DOI: 10.1038/nature12184. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3804423/
12. Koo DH, Molin WT, Saski CA, et al. Extrachromosomal circular DNA-based amplification and gene expression in the evolution of glyphosate resistance in Amaranthus palmeri. Proc Natl Acad Sci USA. 2018;115(13):3332–3337. DOI: 10.1073/pnas.1719354115. https://www.pnas.org/doi/10.1073/pnas.1719354115
13. Mann L, Seibt KM, Weber B, Heitkam T. ECCsplorer: a pipeline to detect extrachromosomal circular DNA (eccDNA) from next‑generation sequencing data. BMC Bioinformatics. 2022;23:40. DOI: 10.1186/s12859-021-04545-2. https://pmc.ncbi.nlm.nih.gov/articles/PMC8760651/
14. Deshpande V, Luebeck J, Nguyen N‑D, Bakhtiari M, Turner KM, Schwab RB, Carter H, Mischel PS, Bafna V. Exploring the landscape of focal amplifications in cancer using AmpliconArchitect. Nature Communications. 2019;10:392. DOI: 10.1038/s41467-018-08200-y. https://www.nature.com/articles/s41467-018-08200-y
15. ENCODE Project Consortium. The ENCODE portal: data resources for functional genomics, including replication-related ChIP-seq. https://www.encodeproject.org

Notes

• All methods and services referenced are for research use only (RUO) and not intended for clinical diagnosis, treatment, or individual health assessment.
• Figures labeled as schematic or illustrative are self-created for educational purposes; replace with your own experimental data for publication.