Golden Gate Overhang Set Designer - Choose Junctions, Not Just Score Them
Pick a high-fidelity set of 4-base fusion sites for an N-part assembly, pinning the overhangs your vector already commits you to.
🔒 Nothing you paste is logged or stored on our servers
The companion to the fidelity scorer: instead of grading a set you already have, this chooses one. Each junction takes one 4-base fusion site, every pair of chosen sites cross-reacts by an amount the published Potapov/Pryor ligation matrices measure, and a one-pot assembly is only as good as its worst junction - so the objective is the weakest link, not the average. Up to about six or seven junctions the search is exhaustive and the answer is provably the best available; above that a node budget stops it and the result says so, because "this is optimal" and "this is the best we found" are different claims. One thing worth knowing before you read the fidelity figure: the ligation matrix is 94% zeros, so a perfect 1.0 usually means the experiment never observed those particular misligations rather than that they cannot happen. Every candidate here is also held above a minimum ligation signal for that reason - an overhang nothing was measured against can score perfectly and barely ligate.
Including any you pin below.
What the destination vector already commits you to.
Say how many junctions your assembly has, and pin any the vector fixes.
How to use the Golden Gate Overhang Designer tool
- 1Enter how many junctions your assembly has, counting any you are pinning.
- 2Pin the overhangs the destination vector already commits you to. They are scored like any other member and simply cannot be swapped out.
- 3Forbid anything you must avoid - a fusion site already used elsewhere in the build, for instance.
- 4Read the weakest-link fidelity and the weakest ligation signal together. A high ratio over a low signal is a set that barely ligates.
Frequently asked questions
How is this different from the fidelity scorer?
The scorer grades a set you bring. This one chooses the set. They share one implementation - the design is re-scored with the scorer itself before it is returned - so the number here is the number that page will give you for the same overhangs.
Why does it refuse palindromes?
A palindromic overhang like GTAC presents the same 4-mer at both ends, so the intended head-to-tail join and the head-to-head misjoin are the same measured lump. It caps at 0.5 fidelity by construction, which is not a set anyone should be handed.
What is the minimum ligation signal for?
It stops the optimiser hunting for overhangs nothing was measured against. The matrix is 94% zeros, so the fidelity ratio hits 1.0 for any pair the experiment never saw misligate - and an unguarded search returns sets like AAAA, AACA, AAGA at a flawless score, where AAAA's own correct-join signal is 187 against a dataset median of 2,694. The floor defaults to that median.
Can I use this set in MoClo?
Not as a drop-in. MoClo and its relatives fix their fusion sites by position, because the overhang encodes which part slot a fragment belongs to. A freely designed set is for an assembly whose junctions you control.
Why does it stop being exhaustive above six or seven junctions?
The space grows about 24-fold per extra pair of junctions - 181 sets examined at two junctions, 5,791 at four, 137,082 at six. Beyond that the node budget stops the search and the result reports exhaustive: false, so you know the answer is the best found rather than the best there is.
More
Related tools
Choose the smallest set of sequencing primers that can tell your candidate molecules apart, and find out which pairs nothing will separate - before you pay for the reads.
Calculate GC%, AT% and per-base composition for DNA or RNA.
Estimate primer Tm, GC% and molecular weight from a sequence.