This restriction site calculator scans a DNA sequence for the recognition sites of 12 workhorse restriction enzymes, reporting counts and positions — the first step of every digest plan and in-silico clone check.
| Enzyme | Site | Overhang | Cuts | Positions (5′ start) |
|---|
Palindromic sites read identically on both strands, so one scan covers both. Digest planning caveats: methylation (Dam/Dcm/CpG) blocks some enzymes on DNA from common strains, star activity relaxes specificity in wrong buffers, and double digests need buffer compatibility — check the manufacturer’s chart before the bench, this page owns the sequence geography only.
Recognition Sites Are Palindromes with Opinions
Most Type II enzymes read 4–8 bp palindromes — the same forward as reverse-complement (our Reverse Complement Calculator shows why) — and cut inside them leaving sticky or blunt ends. Expected frequency: a 6-cutter appears on average every 4⁶ = 4,096 bp of random DNA; an 8-cutter (NotI) every 65,536. Part of our Biology Calculators collection.
Worked Example (Verified by Inspection)
- Default 56 bp demo sequence: EcoRI (GAATTC) cuts at position 4, BamHI (GGATCC) at 12, HindIII (AAGCTT) at 21, ApaI (GGGCCC) at 27, PacI (TTAATTAA) at 39 — a hand-built multiple-cloning-site tour in one line of DNA
- MCS logic in miniature: each of those enzymes cuts exactly once in the demo — unique cutters are cloning gold; zero-cutters (NotI here) are the diagnostic negatives that frame a map
- Frequency reality check: a 4,096 bp plasmid “should” average one EcoRI site — lotteries of sequence mean many plasmids have zero or three; this scan settles arguments the average can’t
- Blunt vs sticky: SmaI’s blunt CCC^GGG products ligate to anything blunt (low efficiency, no direction); sticky ends ligate directionally to matching overhangs — check the overhang column before promising a cloning strategy
Planning Digests Honestly
- Two sites for one enzyme → three fragments on linear DNA (sizes = gaps + ends); circular DNA with n cuts gives n fragments — the map this table starts.
- Sites too close together (< ~10 bp apart or near ends) cut poorly: enzymes need flanking DNA to grip.
- Isocaudomers (different sites, same overhang — BamHI/BglII) ligate cross-compatibly but destroy both sites: a feature in assembly design, a bug in verification digests.
- Methylation: XbaI and others are blocked by overlapping Dam methylation on DNA from common E. coli strains — “site present, enzyme silent” is a classical bench mystery with this mundane answer.
Related Tools
- Reverse Complement Calculator — the other strand, instantly
- GC Content Calculator — composition context for the sequence
- Biology Calculators — the full collection
Frequently Asked Questions
How do I find restriction sites in a DNA sequence?
Scan the sequence (and implicitly its complement — palindromic sites match both) for each enzyme’s recognition pattern, recording 1-based start positions. This calculator checks 12 standard enzymes (EcoRI, BamHI, HindIII, PstI, XbaI, SalI, KpnI, SacI, SmaI, ApaI, NotI, PacI) and tabulates counts and positions.
Why are restriction sites palindromes?
The enzymes are homodimers: two identical protein subunits each read half the site on opposite strands, so the recognition sequence reads the same 5′→3′ on both strands (GAATTC’s complement strand is also GAATTC read forward). That symmetry is what lets one scan find sites on both strands at once.
How often does a 6-cutter cut random DNA?
On average every 4⁶ = 4,096 bp (each position matches a specific 6-mer with probability (1/4)⁶). Real genomes deviate (GC content skews everything — CpG-suppressed vertebrate DNA starves NotI of its GC-rich site), which is exactly why scanning beats averaging for your actual sequence.
What does a “unique cutter” mean?
An enzyme cutting exactly once in your construct — the cloning workhorse, since one cut linearizes a plasmid cleanly for insertion. The demo sequence engineers five unique cutters in 56 bp; real vector multiple-cloning sites stack 10–20 of them deliberately. Zero-cutters matter too: they leave your plasmid untouched for negative-control lanes.
Why didn’t an enzyme cut a site that’s clearly present?
Prime suspects: methylation blocking (Dam/Dcm from the host strain), site too near a DNA end (enzymes need flanking bases), wrong buffer/temperature, or star-activity confusion in reverse. The sequence geography here is necessary truth — enzyme behavior adds its own footnotes from the manufacturer’s data.
Is this restriction site calculator free?
Yes — free, instant, no sign-up.