Sample Submission Guidelines
Experimental Workflow for eccDNA Sequencing: Enrichment, Library Prep, and Common Pitfalls
Extrachromosomal circular DNA (eccDNA) is a "needle in a haystack": rare circular molecules dispersed amid an overwhelming background of linear genomic DNA. Sequencing them reliably requires a disciplined, end-to-end workflow and explicit quality controls. In this practical guide for wet‑lab scientists, we'll walk through a four‑step eccDNA sequencing workflow—high molecular weight (HMW) DNA extraction, linear DNA elimination (eccDNA enrichment), rolling circle amplification (RCA), and library preparation—plus the controls, thresholds, and long‑read validation that turn results into reproducible findings.
If you're new to the topic, start with the concept overview in eccDNA Sequencing Explained.
Step 1: High Molecular Weight (HMW) DNA Extraction
Standard minipreps often fall short for eccDNA discovery because they shear DNA and bias recovery toward smaller fragments. Large circles can be fragile; harsh lysis, repeated vortexing, and aggressive column binding/elution can nick or break them. The goal here is clean, HMW DNA with minimal shearing.
What to aim for
- Intact HMW DNA (broad high‑molecular smear on a low‑voltage 0.8% agarose gel) and accurate dsDNA quantification by Qubit. Avoid relying on NanoDrop alone—protein, phenol, or salt carryover can distort A260/280.
- Gentle lysis and minimal pipetting: think slow, cold, and wide‑bore tips when possible.
- Solid‑phase or bead‑based cleanup that preserves long molecules. Multiple eccDNA studies report robust HMW preservation with bead‑based approaches and confirm integrity via gel/TapeStation profiles.
Suggested inputs and handling
- Recommended input mass: 0.5–2 µg HMW DNA per sample for downstream digestion and RCA, depending on tissue/cell yield.
- Keep EDTA in lysis buffers to inhibit nucleases; avoid prolonged room‑temperature incubations.
- Store intermediate DNA at 4°C for short holds; freeze at −20°C for longer storage. Thaw on ice.
QC checkpoint 1: extraction integrity
- Run 0.8% agarose at low voltage; HMW DNA should present as a broad upper smear with few discrete low‑molecular bands.
- Qubit dsDNA concentration: document yield and purity. Optional TapeStation/Bioanalyzer can capture fragment distribution for your records.
Decision rule
- If gel shows heavy fragmentation or DNA is scarce (<100 ng), consider re‑extracting with gentler lysis and bead‑based cleanup. Do not proceed to exonuclease digestion with clearly degraded DNA; you'll increase false positives and reduce circle recovery.
Evidence context
- Many labs confirm HMW integrity with gels/TapeStation before circle‑seq. Reviews summarize bead‑based methods and integrity checks for eccDNA workflows, e.g., in the Frontiers in Genetics methods review (2024), which discusses extraction choices and downstream enrichment.
Figure 1. eccDNA workflow: HMW extraction → Exo digestion → RCA → library prep → NGS. QC checks: extraction integrity, digestion efficiency, RCA profile, library metrics.
Step 2: Linear DNA Elimination (eccDNA Enrichment)
The cornerstone of an eccDNA sequencing workflow is rigorous removal of linear DNA. Two enzyme families dominate this step:
- Exonuclease V (RecBCD; commonly from NEB): digests linear dsDNA; activity depends on ATP and buffer. It may degrade nicked circles.
- Plasmid‑Safe ATP‑dependent DNase (Epicentre/Lucigen): selectively degrades linear dsDNA while sparing intact closed circular dsDNA; circle‑seq protocols often favor multi‑hour to multi‑day digestions with ATP re‑dosing.
Setups that work in practice
- Exonuclease V: run in 1X NEBuffer 4 with ~1 mM ATP at 37°C; incubation time scales with input mass and background. NEB documentation notes the ATP requirement and selective linear DNA removal; see NEB's Exonuclease V product page and buffer activity guidance.
- Plasmid‑Safe: incubate at 37°C with ATP; for complex samples, extend digestion and replenish enzyme/ATP daily to reduce linear load while avoiding over‑digestion that could nick circles. Circle‑seq implementations using Plasmid‑Safe are summarized in Yu et al. (2023) Circle‑seq overview.
Verification is non‑negotiable
- qPCR of one or two single‑copy genomic loci: compare Ct before vs. after digestion. A substantial Ct increase (e.g., ≥3–4 cycles) indicates linear DNA depletion. Several studies use locus‑specific assays to quantify depletion and confirm enrichment.
- Gel electrophoresis: expect disappearance of prominent linear bands/smear; circular spike‑ins should remain detectable (see the controls section).
Mitochondrial DNA (mtDNA) contamination
- mtDNA can persist and skew results. Many circle‑seq workflows pre‑linearize mtDNA using CRISPR‑Cas9 with sgRNAs at two mtDNA sites before digestion, then confirm removal by PCR; see protocol notes in dos Santos et al. (2023).
QC checkpoint 2: digestion success
- Document qPCR Ct shifts. Capture gel images before/after digestion.
- If Ct shift is <3 cycles and gel shows residual smear, repeat digestion with fresh enzyme and ATP. Keep an eye on circle integrity; avoid prolonged high‑temperature exposure or excessive pipetting.
Decision rule
- Proceed to RCA only after confirming robust linear depletion by qPCR/gel. If digestion remains incomplete, rerun with increased enzyme units, refreshed ATP, or extended incubation. If you suspect circle nicking, switch enzyme or reduce exposure time.
Reference and deeper method selection
- For enzyme choice tradeoffs and control strategies, see Choosing eccDNA Enrichment Methods.
Figure 2. Gel comparison: Before digestion — abundant linear smear; After digestion — depleted linear DNA with retained circular spike‑in (verify by qPCR); Post‑RCA — high‑molecular‑weight concatemer smear indicating successful amplification.
Step 3: Rolling Circle Amplification (RCA) with Phi29
RCA boosts low‑abundance circles so they're detectable in downstream libraries. Phi29 polymerase is the workhorse due to strong strand displacement and high processivity. However, it can produce concatemers and chimeras if conditions aren't carefully managed.
Key parameters
- Primers: use exonuclease‑resistant random primers (e.g., phosphorothioate‑protected) to maintain primer integrity under Phi29's proofreading; see kit guidance such as NEB's Phi29‑XT RCA kit protocol.
- Temperature and duration: WT phi29 typically runs at ~30°C; engineered variants (e.g., phi29‑XT) at 42°C. Limiting RCA to ~2 hours often reduces excessive concatemer formation while delivering sufficient yield, as noted across vendor application notes for multiple displacement amplification.
- Debranching: T7 Endonuclease I treatment or gentle acoustic shearing can reduce branched concatemers before library prep. NEB's RCA guidance suggests debranching for cleaner libraries.
- Inactivation: follow kit protocols for heat or EDTA‑based stops to halt activity.
Controls in RCA
- No‑template control (NTC): should show no amplification; any signal points to contamination.
- Linear negative control: carry a mock digest without circular templates; significant amplification suggests incomplete digestion or primer artifacts.
- Circular spike‑ins: quantify RCA yield and bias across sizes.
QC checkpoint 3: post‑RCA profile
- Gel/TapeStation: expect a high‑molecular smear. Distinct repeated bands may indicate concatemer artifacts; mitigate via reduced reaction time or debranching.
- Quantify RCA yield; note fragment distribution if you plan size selection.
Decision rule
- If RCA produces overly discrete bands or excessive chimeras, shorten incubation, verify primer quality, and consider debranching before fragmentation. Confirm that NTC and linear negatives remain clean.
Context for replication stress inputs
- If your project intentionally perturbs replication (e.g., hydroxyurea treatments), understand how stress impacts circle abundance and RCA inputs. Learn more in Replication Stress and eccDNA.
Figure 3. Phi29 RCA produces concatemeric amplicons from circular templates—mitigate artifacts by shortening reaction time, using exonuclease‑resistant primers, and applying debranching or gentle shearing before library prep.
Step 4: Library Preparation and QC
With enriched, amplified circular templates, you're ready to prepare short‑read libraries (commonly Illumina PE150) for discovery and profiling. Treat library prep as both a construction step and a diagnostic.
Fragmentation
- Target inserts around ~300–400 bp. Over‑fragmentation erodes complexity and can bias junction detection.
- Choose acoustic shearing or enzymatic fragmentation; tune energy/time to hit the insert window.
Adapter ligation and amplification
- Use ligation kits compatible with your fragmentation strategy; avoid excessive PCR cycles—each cycle increases duplication and bias. If inputs are low, consider ligation options optimized for low‑input DNA.
Size selection and cleanup
- SPRI bead ratios: fine‑tune to retain your target insert window. Overly tight selection may discard informative junction‑spanning fragments.
Sequencing depth guidance
- Discovery runs: many circle‑seq studies report per‑sample read counts from tens to hundreds of millions of PE150 reads. Your optimal depth depends on tissue complexity and expected circle abundance; pilot with modest depth, then scale once QC passes. For platform context, see the Next‑Generation Sequencing (Illumina) overview.
- Validation runs: if moving to long‑read confirmation (PacBio HiFi or ONT), depth can focus on candidates rather than global discovery.
Acceptance criteria to record
- Insert size distribution (Bioanalyzer/TapeStation) around 300–400 bp.
- Mapping rate near ~90% with a standard aligner (e.g., BWA‑MEM) indicates good library quality.
- Duplication rates: monitor and keep as low as practical; report along with Q30 metrics and adapter/low‑quality removal details.
Where a turnkey service can help
- For labs preferring a standardized handoff—from digestion through library prep and bioinformatics—Disclosure: CD Genomics is our product and offers eccDNA sequencing and analysis.
Reference acceptance criteria
- Review detailed acceptance criteria, reproducibility metrics, and reporting guidance in Quality Metrics for eccDNA Sequencing.
Controls and Spike‑ins: Designing End-to-End QC
The most significant determinant of reproducibility in an eccDNA sequencing workflow is the control plan. Build three mandatory controls into your pipeline and treat their readouts as project gatekeepers.
- Synthetic circular spike‑ins (paired with a linear version)
- What they do: quantify digestion efficiency and RCA yield, and calibrate detection thresholds.
- How to use: add defined amounts of a circular plasmid or mini‑circle plus a linearized version of the same sequence at extraction or pre‑digestion. After digestion and RCA, compute the log2 ratio of circular:linear by qPCR or read counts.
- What to expect: circle‑seq studies report sizable increases in circular:linear ratios after optimized digestion and RCA, reflecting successful linear depletion and circular enrichment; see examples in Yu et al. (2023) Circle‑seq overview.
- Linear genomic DNA negative control
- What it does: estimates false‑positive background—spurious junction calls or split‑reads that shouldn't appear when circles are absent.
- How to use: process a sample through digestion and RCA that lacks circular templates. Any amplification or junction detection indicates incomplete digestion or RCA/primer artifacts; use this background to set detection thresholds.
- No‑template RCA control (NTC)
- What it does: monitors contamination and primer artifacts in RCA reagents.
- How to use: run an RCA with water instead of DNA. Any product suggests contamination—do not proceed.
eccDNA Sequencing Workflow QC Thresholds and Depth Planning
Detection thresholds
- Short‑read acceptance: require multiple split‑reads supporting a junction (e.g., ≥5 split reads) and consistent coverage around the putative circle. Adjust based on background observed in the linear negative control.
- Long‑read acceptance (for candidates): aim for multiple independent full‑length reads spanning the junction (e.g., ≥2 ONT/PacBio reads), plus split‑read corroboration. Candidate validation approaches with full‑length eccDNA reconstruction are described in methods like FLED; see Li et al. (2023) full‑length eccDNA detection.
Spike‑in calibration
- Use measured recovery and ratio changes to tune sensitivity. If spike‑in recovery is poor, revisit digestion/RCA before sequencing deeper.
Sequencing depth recommendations
- Discovery: begin with ~50–150 million PE150 reads per sample as a pragmatic range, then scale according to control performance and sample complexity. Depth ranges and discovery‑to‑validation transitions are discussed across eccDNA method papers; a general overview of circle‑seq data volumes appears in Jiang et al. (2023).
- Validation: for targeted candidate confirmation, long‑read runs can be modest if they yield full‑length spanning reads; the Complete Overview of Long‑Read Sequencing provides platform tradeoffs useful for depth planning.
Recording QC outcomes
- At minimum, capture: spike‑in ratio changes and recovery percentage, qPCR Ct shifts for single‑copy loci, NTC/negative control outcomes, library insert size distribution, mapping/duplication rates, and minimal supporting reads for accepted circles.
Long-Read Validation Strategy (PacBio/ONT)
Short‑read discovery is powerful, but complex repeats and junction ambiguity benefit from orthogonal long‑read validation. Plan this step early so you can reserve material and avoid re‑extraction.
Candidate selection
- Choose 5–10 putative eccDNAs spanning diverse sizes and repeat content. Favor candidates with strong short‑read support and good spike‑in recovery metrics.
Library strategies
- Option A: sequence RCA concatemers natively on ONT to capture junction‑spanning molecules within long tandem repeats.
- Option B: perform PacBio HiFi libraries on size‑selected DNA to achieve high per‑read accuracy; suitable for confirming precise breakpoints.
- Option C: hybrid plan—use acoustic shearing post‑RCA to tune length distributions suited to the chosen platform.
Inputs and QC
- Verify library length and quality by Bioanalyzer/TapeStation. Ensure minimal adapter dimers and appropriate size distribution.
- Confirm that NTCs and linear negatives remain clean; long‑read signal in negatives indicates contamination or digestion failure.
Interpreting long‑read evidence
- Look for full‑length reads that traverse the junction with high mapping quality and minimal soft‑clipping. Combine with split‑read evidence from short‑read libraries.
- Use graph-based reconstruction or consensus assembly to resolve repeats; require multiple independent molecules before declaring structure.
Platform tradeoffs
- ONT excels at capturing very long concatemers and complex repeats; PacBio HiFi provides excellent per‑read accuracy for high‑confidence breakpoint calls.
Where service context helps
- If you need a turnkey validation path, Complete Overview of Long‑Read Sequencing explains platform tradeoffs. Providers including CD Genomics long‑read sequencing and PacBio SMRT Sequencing pages outline capabilities you can map to your validation plan.
Common Pitfalls and Troubleshooting
High linear background after digestion
- Likely causes: insufficient enzyme units, degraded ATP, or overly short incubation.
- Fixes: repeat digestion with fresh enzyme and ATP; extend incubation or re‑dose daily (for Plasmid‑Safe); verify with qPCR Ct shift and gel.
Mitochondrial contamination
- Likely causes: mtDNA persistence during digestion.
- Fixes: pre‑linearize mtDNA using CRISPR‑Cas9 at two sites before digestion; confirm by PCR.
RCA concatemers and chimeras
- Likely causes: prolonged RCA, primer degradation, or branching.
- Fixes: limit RCA to ~2 h; use exo‑resistant primers; debranch with T7 Endonuclease I; ensure NTC/linear negatives are clean. Vendor protocols such as NEB's Phi29‑XT RCA protocol provide mitigation notes.
Over‑fragmentation during library prep
- Likely causes: excessive shearing time or enzyme exposure.
- Fixes: titrate fragmentation energy/time to hit 300–400 bp inserts; monitor via Bioanalyzer; reduce PCR cycles to preserve diversity.
Low library complexity or high duplication
- Likely causes: low input DNA or over‑amplification.
- Fixes: optimize ligation efficiency; minimize PCR cycles; consider size selection adjustments; re‑evaluate RCA yield and digestion QC.
Bioinformatics Handoff Checklist
Delivering a robust dataset to the analysis team accelerates interpretation and reduces back‑and‑forth. Provide the following minimums:
- Raw read depth (paired‑end and read length) per sample.
- Mapping rate, duplication rate, and Q30 distribution after adapter/quality trimming.
- Insert size distribution from Bioanalyzer/TapeStation.
- Spike‑in recovery metrics: circular:linear log2 ratio changes and percent recovery across sizes.
- qPCR Ct shifts for single‑copy loci before vs. after digestion.
- NTC and linear negative control outcomes (including any reads detected and junction calls, ideally zero).
- Detection thresholds used: minimal split‑reads per junction (short‑read) and minimal full‑length reads (long‑read) for candidate acceptance.
- A shortlist of candidates selected for long‑read validation, with rationale.
Conclusion: Standardize for Reproducible "circle‑seq eccDNA" Results
An eccDNA sequencing workflow lives or dies by its controls. Gentle HMW extraction preserves circles; rigorous exonuclease digestion verified by qPCR/gel removes linear background; carefully tuned RCA amplifies circles without flooding your library with concatemers; and disciplined library QC plus sensible sequencing depth carry reliable signals into analysis. Treat synthetic circular spike‑ins, linear negative controls, and NTCs as mandatory. Plan long‑read validation for representative candidates to lock down junctions and repeat structures.
If you want a standardized, RUO path from enrichment through library prep and reporting, CD Genomics eccDNA sequencing can support project design and deliverables. For deeper platform context, revisit the Next‑Generation Sequencing (Illumina) overview and the Complete Overview of Long‑Read Sequencing.
References:
- Yu X, et al., "Circle‑seq overview and spike‑in quantification" (2023) — methods and spike‑in metrics: Yu et al., 2023 — Circle‑seq (PMC10788182).
- Li Y, et al., "Full‑length eccDNA detection (FLED)" (2023) — long‑read validation and supporting‑read thresholds: Li et al., 2023 — FLED (PMC10632013).
- Jiang Z, et al., "Circle‑seq depth and discovery considerations" (2023) — sequencing depth examples and dataset context: Jiang et al., 2023 (PMC10300174).
- Hong J, et al., "HMW extraction and eccDNA atlas" (2024) — HMW extraction recommendations and mapping rates: Hong et al., 2024 (PMC10973517).
- Hansen K, et al., "Purification comparisons for eccDNA workflows" (2023) — bead‑based vs. alternative purifications: Hansen et al., 2023 (PMC10523981).
- dos Santos R, et al., mtDNA linearization protocol notes (2023) — CRISPR‑Cas9 pre‑linearization recommendations: dos Santos et al., 2023 (PMC10495552).
- Zhuang L, et al., "Circle score and coverage continuity metrics" (2024) — scoring/filter thresholds and continuity metrics: Zhuang et al., 2024 (PMC10948907).
- Chen Q, et al., urinary eccDNA example (2025) — application and detection notes: Chen et al., 2025 (PMC12627897).
- Frontiers in Genetics methods review (2024) — overview of eccDNA extraction and enrichment choices: Frontiers in Genetics methods review (2024).
- Qiu et al., "Extrachromosomal circular DNA as a novel biomarker" (PMC) — broader context and review: Qiu et al. (review) (PMC).
- Ye et al., PSD usage and methodological notes (2023) — example PSD applications in Circle‑seq: Ye et al., 2023 (PMC10670553).
- Lin et al., PSD and enrichment descriptions (2024) — additional PSD usage examples: Lin et al., 2024 (PMC11606223).
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.
