Amplicon Deep Sequencing (CRISPResso2) for Indel Spectrum Quantification at an Edited Locus

Three biological samples (edited bulk population, unedited parental control, no-template control) each with three technical PCR replicates. Each amplicon designed to place the cut site asymmetrically (~1/3 from one end) within a 250-300 bp amplicon for paired-end 2x150 read overlap. Target >5,000 aligned reads per amplicon (typically 10,000-50,000). Unedited control defines the background substitution/indel noise floor that sets the limit of detection.

BSL-2 if amplifying from human-cell-derived gDNA; purified gDNA itself is low-risk. Standard PPE (gloves, lab coat, eye protection). Maintain pre-PCR and post-PCR workspace separation to prevent amplicon carryover contamination; use filter tips and dedicated pipettes. Ethidium-free dyes preferred for any gel checks; if SYBR/EtBr used, treat as a mutagen and dispose appropriately. No special biohazard beyond cell-line handling.

Unedited parental gDNA processed identically defines the PCR/sequencing error background per position (subtracted as noise). No-template control (water) in both PCR rounds detects contamination. A spike-in of a synthetic known-indel allele at 1% can serve as a positive quantification standard. Include a previously characterized edited sample as a run-to-run positive control. PhiX provides base-diversity/calibration control for the low-complexity amplicon library.

Edited population should show a clear indel frequency (e.g., 40-90%) with a characteristic spectrum dominated by 1 bp insertions and small deletions; frameshift fraction typically 60-70% of edits for a non-MMEJ-biased site. Unedited control indel background should be <0.5%. Limit of detection ~0.1-0.5% with >10,000x coverage. Technical replicates should agree within +/-3% absolute indel frequency.

• Genomic DNA from edited and unedited cells, 50-100 ng/µL, A260/280 ~1.8
• High-fidelity polymerase (e.g., NEB Q5 Hot Start 2X Master Mix M0494)
• Round-1 locus primers with universal Illumina partial adapter tails (read1: 5'-TCGTCGGCAGCGTCAGATGTGTATAAGAGACAG-[locus-F]; read2: 5'-GTCTCGTGGGCTCGGAGATGTGTATAAGAGACAG-[locus-R])
• Round-2 indexing primers (Nextera i5/i7, e.g., IDT for Illumina UD indexes)
• AMPure XP beads (Beckman A63881) or equivalent SPRI
• Qubit dsDNA HS Assay Kit (Q32854)
• High-Sensitivity DNA chip (Agilent Bioanalyzer 2100) or TapeStation D1000
• KAPA Library Quantification Kit (Roche) for qPCR pooling
• Illumina MiSeq Reagent Kit v2 (300 cycles) or iSeq 100
• PhiX Control v3 (Illumina FC-110-3001)

Establish a robust two-round PCR (tagmentation-free) amplicon-seq workflow to measure indel frequency, frameshift fraction, and the precise distribution of insertion/deletion alleles at a CRISPR-edited locus, with quantitative limits of detection below 1% editing.

1. Round-1 PCR: 20 ng gDNA, 0.5 µM each tailed locus primer, Q5 2X MM, 25 µL; 98 C 30 s; 25 cycles of 98 C 10 s / 62 C 20 s / 72 C 20 s; 72 C 2 min.
2. SPRI clean-up at 0.9x bead ratio; elute in 20 µL EB.
3. Round-2 indexing PCR: 5 µL round-1 product, 0.5 µM each unique i5/i7 index primer, Q5 2X MM, 25 µL; 98 C 30 s; 8 cycles of 98 C 10 s / 62 C 20 s / 72 C 20 s; 72 C 2 min.
4. SPRI clean-up at 0.9x; elute in 25 µL.
5. Quantify each library by Qubit HS; size-verify (~380-430 bp) on TapeStation.
6. Normalize and pool to equimolar; qPCR-quantify the pool with KAPA.
7. Denature, spike 10-15% PhiX (low-diversity amplicon), load MiSeq at 8-10 pM, 2x150 paired-end.
8. Demultiplex to FASTQ; run CRISPResso2 per amplicon with the reference and sgRNA sequences.

Independent variable: editing status of the input gDNA (edited vs unedited). Dependent variables: total indel frequency (%), frameshift fraction (%), per-allele indel size distribution, substitution rate. Controlled variables: input gDNA amount (20 ng), polymerase and cycle number (round-1 = 25, round-2 = 8), amplicon design and primer set, sequencing depth target (>5,000x), CRISPResso2 quantification window (default +/-15 bp around cut site), minimum read quality (Q30).

Targeted amplicon deep sequencing at >5,000x coverage will resolve the full indel spectrum at the edited site, detect editing events present at <1% allele frequency that Sanger/TIDE miss, and yield indel frequencies reproducible within +/-3% across technical PCR replicates.

Process FASTQ with CRISPResso2 specifying amplicon reference, guide sequence, and quantification window; merge overlapping read pairs. Report modified-read %, indel size histogram, frameshift vs in-frame split, and substitution profile. Subtract the matched unedited-control background frequency per indel class. Aggregate biological replicates; visualize allele frequency tables and indel-size plots. Confirm coverage sufficiency (>5,000 aligned reads) before reporting; flag amplicons with <1,000 reads as failed.

Low or no round-1 product: re-design primers (Tm ~62 C, avoid SNPs/repeats), increase gDNA to 40 ng, or drop anneal to 60 C. Primer-dimer dominating the TapeStation trace: increase SPRI to 0.8x and verify primer specificity. Uneven coverage across pooled samples: re-quantify by qPCR (KAPA) rather than fluorometry alone and re-pool. CRISPResso2 reports high substitutions in unedited control: low base quality or amplicon SNP — raise the Q-score filter and confirm reference. Amplicon too long for read overlap: redesign to <300 bp so 2x150 reads merge across the cut site.

Editing frequency reported as mean +/- SD across three biological replicates (technical replicates averaged first). Compare edited vs unedited indel frequency by unpaired two-tailed t-test, alpha = 0.05. When comparing multiple guides/conditions, use one-way ANOVA with Tukey HSD correction. Estimate the limit of detection as mean unedited background + 3 SD. With three biological replicates and the large edited-vs-background effect, power exceeds 0.9.