Base & prime editing
A cytidine deaminase on a Cas9 nickase converts C to uracil, read as T on copying, giving C•G → T•A. Used to install stop codons, correct T→C point mutations and model variants. The design problem is bystander Cs inside the window, and the quality problem is product purity, which is why a uracil glycosylase inhibitor is part of the construct.
Our ruleSimulations prepare you for the bench but don't replace hands-on practice.
Cytosine base editors
A cytidine deaminase on a Cas9 nickase converts C to uracil, read as T on copying, giving C•G → T•A. Used to install stop codons, correct T→C point mutations and model variants. The design problem is bystander Cs inside the window, and the quality problem is product purity, which is why a uracil glycosylase inhibitor is part of the construct.
How it differs from the standard method
- Sequence context matters: a C preceded by G is edited less efficiently by the commonly used deaminases.
- Without UGI, edited reads come back as a mixture of T, G and A at the target position.
- Guide-independent deamination is a documented off-target mode, so a non-targeting guide belongs in every experiment.
1Editing without breaking both strands
Cas9 nuclease editing works by breaking the DNA and letting the cell repair it. That is why its products are a mixture of indels, why precise changes need a donor template and homology-directed repair, and why it works poorly in cells that are not dividing.
2Base editors change one letter, within a window
A base editor is a deaminase tethered to a Cas9 nickase. Once the guide has opened the DNA, the deaminase acts on whichever bases of the exposed strand it can reach.
C•G → T•A
Cytidine deaminase plus a uracil glycosylase inhibitor. Without the inhibitor the cell excises the uracil and repairs it unpredictably, so some edited reads come back as G or A instead of T.
A•T → G•C
An evolved adenosine deaminase, which does not occur naturally on DNA. Generally cleaner than CBEs, with a narrower effective window and less sequence-context preference.
Any short change
A reverse transcriptase writing from a template on the guide itself. Substitutions, small insertions and small deletions, at lower efficiency and with more design to get right.
3Prime editing writes from a template it carries
A prime editor nicks one strand and then copies new sequence into the nick from an RNA template built into an extended guide, the pegRNA. Two lengths on that pegRNA decide whether it works: the primer-binding site, which is what the nicked DNA end anneals to before copying starts, and the reverse-transcriptase template, which carries the edit plus enough homology to anneal back into the genome afterwards.
4How these fail, and how you would know
- Bystander editing: the right base changed, but so did its neighbours, so most edited alleles carry extra changes. Only visible if you quantify every position in the window, not just the target.
- Low product purity in CBEs: the target position is edited but reads back as a mix of T, G and A, because uracil was excised rather than copied through.
- Guide-independent deamination: a base editor can act on DNA it was never aimed at, which a non-targeting-guide control is what reveals.
- Sequencing artefacts masquerading as edits: an error-prone or over-cycled amplicon PCR puts roughly 1% apparent substitutions everywhere, which swamps a small real edit.
- A primer that overlaps the window amplifies the unedited allele better than the edited one, so the edit you made is the one you fail to measure.
Key terms
Practise these as flashcards →The few protospacer positions a tethered deaminase can reach, typically around positions 4–8 from the PAM-distal end. It is a property of the editor, not something you aim.
A second C or A inside the window that was edited along with the target, putting extra changes on the same allele.
The fraction of edited reads carrying the intended base rather than one of the other possible outcomes at that position.
Uracil glycosylase inhibitor, included in cytosine base editors to stop the cell excising the uracil intermediate before it is copied through.
A prime-editing guide extended with a primer-binding site and a reverse-transcriptase template carrying the edit.
The primer-binding site the nicked strand anneals to, and the template copied into the nick. Their lengths are the main design variables in prime editing.
Nicking strategies: no second nick, a nick on the other strand, or a nick that only occurs once the edit is installed. PE3 raises efficiency and indels; PE3b raises efficiency with fewer indels.
Reading many copies of a short PCR product over the edited site to get the percentage of alleles carrying each outcome.
When to use it
- You need one defined base change and cannot accept a mixture of insertions and deletions.
- The cells are not dividing, so homology-directed repair is unavailable and a Cas9 knock-in would not work.
- A double-strand break at this locus is unacceptable, for example because the cells are destined for therapeutic use.
- You need a short insertion or deletion of defined sequence, which is prime editing's territory rather than a base editor's.
Compared with alternatives
Reach for nuclease editing when you want a gene disrupted rather than a base changed, or when you need a large insertion.
When the question is which gene matters rather than what one base does, a pooled screen answers it at scale.
The equivalent job on a plasmid, where you can simply amplify the whole molecule and sequence the product.
Older nucleases and recombinase systems, which reach sequences a PAM-constrained editor cannot.
Sub-techniques
Choosing one here switches the whole page, including the protocol and the lab, to it.
A cytidine deaminase on a Cas9 nickase converts C to uracil, read as T on copying, giving C•G → T•A. Used to install stop codons, correct T→C point mutations and model variants. The design problem is bystander Cs inside the window, and the quality problem is product purity, which is why a uracil glycosylase inhibitor is part of the construct.
- Sequence context matters: a C preceded by G is edited less efficiently by the commonly used deaminases.
- Without UGI, edited reads come back as a mixture of T, G and A at the target position.
- Guide-independent deamination is a documented off-target mode, so a non-targeting guide belongs in every experiment.