Paste DNA. It is transcribed to mRNA, cut into triplets, and each triplet is looked up in the genetic code — the same 64-entry table every cell reads. Shift the reading frame by one base and watch a clean protein turn into nonsense.
Line numbers, spaces and > header lines are ignored. Lowercase is fine.
N and the other ambiguity letters are resolved whenever every possibility gives the same
amino acid. Anything else is flagged and read as unknown.
Click a bead to inspect its codon — or focus the chain and walk it with ← / →. Residues outside the start-to-stop stretch are drawn hollow.
Shown in mRNA letters, grouped by first base; rows are the second base, columns the third. The selected codon is ringed. Notice how often a whole row of four says the same thing — that redundancy is why many single-base typos change nothing.
The whole translator is one string: FFLLSSSSYY**CC*W… — 64 amino-acid letters in
codon order, unpacked into a lookup map at load. Every triplet is one map read.
Transcription here is just T→U. Real cells copy the template strand, so the mRNA matches the coding strand you pasted — which is why the letters look unchanged apart from the U.
Reading frame. Nothing in the sequence says where to start counting. Frames 2 and 3 shift the cut by one and two bases, and almost always shatter into stop codons — that is the point.
Ambiguity. A codon containing N (or R, Y, S, W, K, M, B, D, H, V) is expanded
into every base it could be. If all of them land on the same amino acid — TCN is always
serine — the residue is called. If not, it is X.
Stop codon nicknames are the old lab ones: TAA ochre, TAG amber, TGA opal.
What this does not do: introns and splicing, alternative codon tables, signal-peptide cleavage, folding, or anything else that happens after the ribosome lets go. The insulin precursor loaded here becomes actual insulin only after a lot of cutting this page knows nothing about.