Prime editing nicks only one DNA strand, then uses a prime-editing guide RNA (pegRNA) to template new genetic information directly onto the exposed 3′ end — no double-strand break, no donor template.
- Nick + prime. A Cas9 nickase (H840A) nicks the PAM strand. The pegRNA's primer-binding site (PBS) hybridizes to the freed 3′ end, priming reverse transcriptase.
- RT extension. Reverse transcriptase copies the pegRNA's RT template, extending a new 3′ flap that carries the edit — displacing the original, unedited 5′ flap.
- Flap equilibrium. Cellular flap endonucleases (FEN1) and ligases resolve the branched intermediate. Preferential excision of the 5′ flap and ligation of the 3′ flap fixes the edit; the reverse outcome reverts to wild type.
- PE3. Nicking the *non*-edited strand biases mismatch repair toward using the edited strand as template, roughly tripling efficiency — at the cost of occasional simultaneous double-strand nicks that repair as small indels.
This simulator scores each cycle with a simplified but literature-consistent model:
p_hybrid = exp( -(PBS-13)^2 / 32 ) // Tm optimum near 13 nt
p_flap = exp( -RT_len / 22 ) // longer edits excise less efficiently
p_edit = min(0.95, 0.9 · p_hybrid · p_flap · (PE3 ? 1.55 : 1))
p_indel = min(0.35, 0.015 + 0.0025·RT_len + (PE3 ? 0.045 : 0))
p_revert = 1 − p_edit − p_indel
Real prime editors (PE2/PE3, Anzalone et al. 2019) show the same qualitative trends: an optimal PBS length around 12–14 nt, falling efficiency for longer insertions/edits, and a PE3 efficiency boost paired with a modest indel cost — the exact percentages here are illustrative, not measured values.