The Genetic Code: From Gene to Protein

Three bases name one amino acid. That lookup table is shared by every organism on Earth, and it is the closest thing biology has to a universal constant.

lesson 3 of 4 in this unit

Builds on: 2.1 DNA: Four Letters, One Molecule0.3 Enzymes: Why Biology Happens at All

Why proteins are the point

DNA does almost nothing directly. It is an archive. The molecules that actually do the work are proteins: enzymes that catalyse reactions, channels that ferry things across membranes, antibodies, haemoglobin, keratin, muscle fibres. A gene is worth having because it specifies a protein.

A protein is a chain of amino acids, of which there are twenty. The chain folds into a shape determined by its sequence, and β€” as the enzyme lesson established β€” the shape is the function. So the entire problem is: how do you get from an order of four bases to an order of twenty amino acids?

Three letters at a time

One base per amino acid gives 4 possibilities β€” not enough. Two bases give 4Β² = 16 β€” still short of 20. Three bases give 4Β³ = 64, which is more than enough. Nature uses three, and a triplet is called a codon.

Sixty-four codons for twenty amino acids means the code is degenerate: most amino acids have several codons. The redundancy is not random β€” it sits mostly in the third base. GGU, GGC, GGA and GGG all mean glycine, so the third letter can be anything at all. This is called wobble, and it is a genuine error-tolerance feature: many single-base mutations change a codon without changing the amino acid, and are therefore silent.

Three codons name no amino acid at all β€” they are STOP signals:

UAA Β· UAG Β· UGAU Are Away Β· U Are Gone Β· U Go Away β€” the classic hook, and it works

And AUG does double duty: it means methionine, and it is the START codon that sets the reading frame. Where translation begins decides how every subsequent triplet is grouped, which is why frameshifts are so destructive.

The two steps

Transcription, in the nucleus. One gene is copied into mRNA β€” single-stranded, and with U (uracil) wherever DNA would use T. The working copy leaves; the archive never does. This is the same reason a library lends photocopies of a rare manuscript.

Translation, at the ribosome. The ribosome reads the mRNA three bases at a time. For each codon, a tRNA molecule carrying the matching amino acid docks, its anticodon pairing with the codon, and the amino acid is added to the growing chain. At a stop codon the ribosome releases the finished protein.

DNA β†’ (transcription) β†’ mRNA β†’ (translation) β†’ proteinthe central dogma, in its everyday form
One code, all of life

The same 64 codons mean the same amino acids in bacteria, yeast, oak trees and you, with only tiny exceptions. That is why a human insulin gene inserted into E. coli produces human insulin β€” which is how insulin has been manufactured since 1982. The universality is also the strongest single piece of evidence that everything alive shares one ancestor: an arbitrary code, frozen in before the lineages split.

βš—οΈ Lab β€” Transcribe & Translate

A DNA template at the top, a protein at the bottom. Step through it codon by codon.

  • Start with the default gene: TAC transcribes to AUG, so translation begins with methionine.
  • Drag the codon slider one triplet at a time and watch each tRNA deliver its amino acid.
  • Change a third base β€” say GGA to GGG. The protein does not change. That is wobble.
3
mRNA
AUGGGCCCUAAUUGA
Protein
MGPN
Codons
5
Reading frame
fixed by the start codon

Check your understanding

1. Why does the genetic code use three bases per amino acid rather than two?

2. What does it mean that the genetic code is 'degenerate'?

3. Which codon is the START signal, and which amino acid does it also specify?

4. What is transcribed from a DNA template strand reading TAC?

5. Why can a human gene inserted into a bacterium produce a working human protein?