Designing a CRISPR Knockout Experiment End to End: From Guide RNA to Verified Edit

The workflow, in order

A knockout project has five stages, and the order matters. Skipping ahead — editing cells before genotyping primers exist, for instance — turns a quick PCR check into a redesign cycle on cells you already spent weeks generating.

1. Design and rank candidate guide RNAs against the target exon.
2. Design and validate genotyping primers around the expected cut site, before touching the bench.
3. Transfect or nucleofect, select, and expand clones or a bulk pool.
4. PCR across the cut site and sequence the product to check for editing.
5. Compare the edited sequence to the wild-type reference to characterize the change.

Step 1: pick a guide RNA against the target exon

Guide selection starts with scanning the target region for valid PAM sites for whichever nuclease you are using — SpCas9, SaCas9 or Cas12a — then ranking candidates by on-target score and checking the top ones for off-target hits elsewhere in the genome. That process, including the PAM rules and scoring logic for each enzyme, is covered in full in the dedicated guide to CRISPR guide RNA design. SeqBench's CRISPR gRNA Designer runs the PAM scan and on-target scoring on a pasted sequence and returns ranked candidates with PAMs and scores, a shortlist to carry into off-target checking before you settle on which guides to order.

Step 2: design your genotyping primers before you edit anything

This is the step people skip, and the one that costs the most time when skipped. Before you transfect a single cell, design the PCR primers you will use to screen edited clones later, flanking the cut site closely enough that the edit you expect changes something measurable.

• Size-shift design: place primers so the anticipated indel changes amplicon length enough to resolve on a gel — a few bases is invisible on standard agarose, so plan for a larger indel or a higher-resolution method.
• PCR-RFLP design: if the cut site overlaps or is expected to disrupt a restriction site (or a repair template introduces one), primers spanning that site let you screen by digest instead of, or alongside, size shift.
• Specificity and product size: confirm each primer pair amplifies a single, correctly sized product from the target locus with no obvious off-target product before ordering.

Checking genotyping primers before you order them

That last check is what an in silico PCR run gives you: enter the template region and candidate primer pair and confirm the predicted product, its size, and its position before committing to synthesis. Redesigning genotyping primers after you already have edited clones wastes an entire round of screening.

Step 3: screen clones after selection

Once transfection, selection, and expansion are done, screen individual clones or the bulk pool by PCR across the cut site using the genotyping primers from step 2. A bulk pool shows whether editing happened at all and roughly how efficiently; individual clones give you a defined, homogeneous genotype rather than a mixture of edited and unedited alleles.

Step 4: sequence across the cut site and read the trace carefully

PCR product in hand, the next step is Sanger sequencing across the cut site. This is where otherwise successful edits get misread. CRISPR acts on a diploid or polyploid locus, or on a mixed cell population, so a genuinely edited but not yet clonally pure sample commonly produces heterozygous or mosaic outcomes at that position.

In practice that shows up as overlapping double peaks in the trace right at the cut site — a signature of editing having happened in a mixed population, not automatically a failed or noisy read. The guide on verifying a clone by Sanger sequencing covers how to tell a genuine mixed-template double peak apart from ordinary trace noise, and how a downstream indel throws every subsequent position out of register in a way a simple mismatch does not.

Step 5: characterize exactly what changed

Once you have a clean single read from a single-cell-derived clone, where mosaicism is no longer a confound, compare it systematically against the expected wild-type reference. What matters is not just whether it matches, but exactly how it differs: a substitution, an insertion, or a deletion, and the size of that indel.

Size and frame turn a sequence difference into a functional result. A frameshifting indel — one whose length is not a multiple of three — shifts the reading frame downstream of the cut and typically produces a premature stop codon, the outcome you actually want from a knockout. An in-frame deletion of a multiple of three bases removes codons without shifting the frame, which can leave a truncated but partially or fully functional protein, not a knockout, even though the sequence clearly changed. The Variant Comparator aligns your edited sequence to the reference and lists the substitutions, insertions and deletions with their effects, the systematic version of this check rather than eyeballing an alignment.

Where SeqBench fits across the workflow

The single highest-leverage step in this process is designing the genotyping PCR primers before doing the edit, not after. Good primer placement — flanking the cut site with a predictable, already-checked product size — turns downstream screening into a fast, unambiguous check instead of a guessing game run on cells you cannot easily regenerate. The CRISPR gRNA Designer gets you from a target exon to a ranked shortlist of guides; an in silico PCR check confirms your genotyping primers' product and specificity before ordering; Sanger vs Reference turns a sequencing trace into a pass or needs-review report once you have a clone to check; and the Variant Comparator lays out exactly what changed once you have a clean read to characterize. Used in that order, the four tools cover the project from cut-site selection to a verified, characterized knockout.

Related references

Related tools

Related guides

• How to Verify a Clone with Sanger Sequencing →
• How to Design a CRISPR Guide RNA (gRNA): PAM, Scoring and Off-Targets →
• In Silico PCR: How to Predict a PCR Product Before You Run It →