ToolsAbout

Silent mutation finder

Single-base changes that add or remove a restriction site and leave the protein as it was.

Result
How to add a restriction site without changing the protein
Paste the coding sequence, set the reading frame, and the Add a site tab looks at every window of the sequence that is one base away from a recognition sequence. For each of them it tries the base changes that would complete the site, keeps only the ones whose codon still spells the same amino acid, and lists them. The Remove a site tab does the opposite: it finds the sites that are already there and the single silent changes that break them.
…GAG CTG… (Glu Leu) → …GAA CTG… (Glu Leu)  one base, same protein, new site
Both strands are searched, so a site that reads on the bottom strand is found as well. Only the first record of a FASTA file is used.
Worked example
"Try an example" loads the first 150 bases of the EGFP coding sequence, 50 codons in frame 1. On Add a site the tool reports 44 single-base silent changes over29 enzymes, 27 of which have no other site in the fragment. The first row isEarI: its CTCTTC site is completed by changing base 18 from G to A, so codon 6 goes from GAG to GAA, both glutamate. On Remove a site the same sequence gives27 changes that break one of the 6 sites already present, starting with the AleI site at 27, where base 27 goes from C to T and codon 9 from TTC to TTT, both phenylalanine.
Why the third base of a codon does most of the work
The genetic code is degenerate. Leucine, serine and arginine have six codons each, eight more amino acids have four, and nine have two. In the four-codon families the third base is free: GCA, GCC, GCG and GCT all mean alanine, so any of the four can be chosen to suit a recognition sequence. In the two-codon families the third base can swap between the two pyrimidines or between the two purines. Only methionine and tryptophan have a single codon and cannot be touched at all.
That is why a site can usually be placed somewhere near the position you want, but almost never exactly at it. The full list of codons is on thecodon table page, and theDNA to protein translator will confirm that a sequence you have edited still gives the protein you expect.
Reading the result
  • Base is the 1-based position of the single base to change, counted along the sequence as you pasted it, and the change itself.
  • Codon is the codon number in the chosen frame, so codon 1 is the first complete codon after the frame offset.
  • Change gives the old and the new codon with the amino acid they both spell.
  • Also cuts on the Add tab counts the sites that enzyme already has in the sequence. A zero there means the new site will be unique, which is what you want for screening.
  • A change is listed once for each enzyme it suits, so one base change can appear on several rows.
Turning a candidate into a primer
  1. Pick a change whose enzyme has no other site in the construct, and whose new site is clear of the ends of the insert, so the diagnostic fragments differ in size.
  2. Design a mutagenic primer with the change near the middle and 12 to 18 matching bases on each side. The Tm calculator gives the annealing temperature.
  3. Amplify, transform, and screen by digest: the clones that carry the change give the extra band. Check the cut pattern first with therestriction site finder.
  4. Confirm one positive clone by sequencing.
If the goal is to remove a site so that an enzyme cuts only once in the whole plasmid, remember to check the backbone as well, not just the insert.
Limits
  • Only the standard genetic code is used. A mitochondrial or ciliate sequence needs a different table.
  • Codon usage is ignored. A silent change can still swap a common codon for a rare one and slow translation down, which matters in a high-level expression construct. Therare codon analyser will flag that.
  • Only single-base changes are searched, and the sequence is treated as one uninterrupted coding region with no introns.
  • An ambiguous base such as N is never counted as matching a recognition sequence.
Frequently asked questions
What is a silent mutation?
A base change that leaves the protein unchanged, because the new codon still spells the same amino acid. Most amino acids have two, four or six codons, and they usually differ only in the third base, so there is room to edit the DNA without touching the protein. Silent changes are the standard way to put a diagnostic restriction site into a coding sequence.
Why would I add a restriction site to a coding sequence?
To tell clones apart. A silent site that is unique in the construct turns screening into a single digest: the right clone gives an extra band, the wrong one does not. The same trick marks a point mutation, so a site-directed mutagenesis product can be checked without sequencing every colony.
Why would I remove a site?
Usually to free an enzyme for cloning. If the site you want to use in the polylinker also occurs inside your insert, the digest cuts the insert in two. Taking the internal site out with a silent change makes the enzyme a single cutter again. Removing internal sites is also part of domesticating a part for Golden Gate assembly, where BsaI and BsmBI sites have to be unique.
Which reading frame should I choose?
Frame 1 if the sequence starts at the A of the ATG, which is the usual case for a coding sequence pulled out of a database. Choose frame 2 or 3 if the first one or two bases are left over from a cloning scar or a partial codon. The frame decides which triplets are read, so the wrong frame gives the wrong amino acids and the wrong answers.
Does "Any enzyme" really test every enzyme?
It tests the 127 enzymes whose recognition sequence has six or more defined bases, because a four-base site turns up by chance every few hundred bases and is useless for screening. Pick a named enzyme from the list to include the four and five-base cutters, or to work with one enzyme you already have in the freezer.
Is one base change always enough?
No, and the tool only looks for changes of a single base. A site that needs two changes can often still be built, but it needs a longer mutagenic primer and each change must be silent on its own, so it is worth checking by hand. If nothing is found, try a different enzyme or a different part of the sequence.