TuringDNA

Start here

Quickstart: design your first guides

From a gene symbol to a ranked list of guides with ready-to-order cloning oligos, in three steps and about two minutes. Uses the defaults throughout.

Updated 31 July 2026

All documentation

Start here

Overview Quickstart Core concepts

Guides

Turing (the agent) CRISPR knockout Base editing Primer check Plasmid build Variant library

Reference

API Limits

Help

Troubleshooting
On this page
  1. Step 1 — Give it a sequence
  2. Step 2 — Take the defaults
  3. Step 3 — Read the table, then export
  4. What to read next

This is the shortest honest path from nothing to a result you could act on. It uses the defaults at every step, because the defaults are already the right answer for most first runs.

What you need

Nothing, to read the results. A free account to run the tool — CRISPR, primer analysis and the plasmid editor all require sign-in, and only the directed-evolution and agent paths get an anonymous trial.

Step 1 — Give it a sequence

Open the CRISPR designer. You have three ways in, and they are interchangeable:

  • Paste DNA. Raw sequence or FASTA. Headers and whitespace are stripped and anything outside ACGTN is removed, so a copy-paste out of a GenBank record works without cleaning it up first.
  • Type a gene symbol — TP53, GAL1 — and press Fetch. The engine resolves it against Ensembl for human and mouse, and UniProt plus EMBL for everything else.
  • Paste an accession. ENST…, ENSG…, NM_…, NR_…, XM_… and XR_… are all recognised.

If you just want to see it work, press Try an example. That loads the 720 bp EGFP coding sequence — a clean CDS that starts ATG and ends TAA:

Input
ATGGTGAGCAAGGGCGAGGAGCTGTTCACCGGGGTGGTGCCCATCCTGGTCGAGCTGGACGGCGACGTAAACGGCCACAAG
TTCAGCGTGTCCGGCGAGGGCGAGGGCGATGCCACCTACGGCAAGCTGACCCTGAAGTTCATCTGCACCACCGGCAAGCTG
CCCGTGCCCTGGCCCACCCTCGTGACCACCCTGACCTACGGCGTGCAGTGCTTCAGCCGCTACCCCGACCACATGAAGCAG

EGFP is a good first target precisely because it is a transgene: screening it against a host genome asks a real question — does my GFP guide also cut something in the host? — rather than a circular one.

Step 2 — Take the defaults

There are five settings. Four of them you can ignore on a first run.

Setting Default Change it when
Nuclease SpCas9 (20 nt spacer, NGG PAM) You need TTTV-adjacent targets or multiplexing — then pick Cas12a
Mode Knockout You want a single-base change — then pick Base edit
Organism unset You want genome-wide off-target screening. Pick one.
Gene symbol empty You want exon context on each guide
Vector px459_v2 for Cas9, py094 for Cas12a Your lab clones into something else

The one worth setting on your first run is Organism. Leave it unset and only the sequence you pasted is screened for off-targets. Set it to E. coli and your top-ranked guides are screened against the whole 4.6 Mb genome, which takes seconds.

The first run against a new organism is slow

Genome indexes are built on first use — seconds for E. coli, around three minutes for human. Until the index finishes, guides come back with no genome column and a “check again shortly” banner. That is the index building, not a failure. Run it once, wait, run it again.

Step 3 — Read the table, then export

Press Design guides. You get every valid guide in the sequence, ranked. The columns that matter on a first pass:

Column What it means What you want
On-target Doench-style activity heuristic Higher. Use it to rank, not as an absolute efficiency
Input CFD Worst off-target inside your pasted sequence Lower
Genome off Worst off-target in the reference genome Lower, and check the count
Frameshift % Predicted out-of-frame outcomes Higher, for a knockout
Oligos Sense/antisense pair with your vector’s overhangs Copy straight into an order

Then either Export the whole table as Excel or CSV, or hit Order on a single row to copy that guide’s oligo pair to the clipboard.

Research use only

Nothing here is validated for clinical, diagnostic, therapeutic, prophylactic, food, feed or cosmetic use. Every score on the page is a prediction, and every export carries that statement.

What to read next

  • The full knockout guide covers exon context, vector choice, prior-art screening and logging what actually happened at the bench.
  • Core concepts explains what CFD, PAM geometry and the activity window actually are, if the table above used words you had to skip.
  • Limits is worth reading before you trust a number.