How DNA Makes a Protein: Transcription and Translation

A gene is a recipe written in four letters. See how a cell copies it into mRNA, reads it three letters at a time, builds a protein, and what a one-letter mistake does.

DNA doesn't make anything itself. It's a set of instructions, locked away in the nucleus. To build a protein, the cell first copies one gene into a smaller, portable molecule called mRNA (that's transcription), then a ribosome reads the mRNA three letters at a time and joins amino acids together in the matching order (that's translation). The order of the amino acids decides how the protein folds, and its shape decides its job. That's how DNA makes a protein, and the whole two-step process is called protein synthesis.

The four-letter code

DNA is two long strands twisted into a double helix. Each strand is a chain of bases, and there are only four: A (adenine), T (thymine), C (cytosine) and G (guanine). Across the middle, the bases pair up by a strict rule:

A always pairs with T (held by two hydrogen bonds)
C always pairs with G (held by three)

So if one strand reads ATGGTGCAC, the other must read TACCACGTG. Each strand carries all the information needed to rebuild the other, which is exactly how DNA gets copied when a cell divides.

Step 1: transcription (DNA → mRNA)

A gene might be thousands of bases long, and it can't leave the nucleus. So the cell makes a working copy.

An enzyme called RNA polymerase unzips a short stretch of the double helix.
It reads one of the strands (the template strand) and builds a matching mRNA strand, base by base.
RNA uses U (uracil) instead of T, so wherever the DNA has A, the mRNA gets U.
The finished mRNA peels away, the DNA zips back up, and the mRNA leaves the nucleus.

The neat result is that the mRNA reads the same as the other DNA strand (the coding strand), just with U in place of T. ATGGTGCAC in the gene becomes AUGGUGCAC in the mRNA.

Step 2: translation (mRNA → protein)

Out in the cytoplasm, a ribosome clamps onto the mRNA and reads it in groups of three bases. Each group is a codon, and each codon stands for one amino acid.

Reading starts at the first AUG, which codes for methionine. That's why almost every protein starts with it.
For each codon, a tRNA molecule brings the matching amino acid and the ribosome joins it to the growing chain.
Reading stops at a stop codon: UAA, UAG or UGA. The chain is released and folds into its final shape.

So AUG GUG CAC CUG ACU CCU GAG reads as Met–Val–His–Leu–Thr–Pro–Glu. Those are the first seven amino acids of haemoglobin, the protein that carries oxygen in your red blood cells.

Why three letters? With four bases, codons of one letter give only 4 combinations and codons of two letters only 16. That's not enough for 20 amino acids. Three letters give 64, plenty, so several codons can share the same amino acid.

When one letter changes: types of mutation

A mutation is a change in the DNA sequence. What it does depends on the change:

Type	What happens	Effect on the protein
Silent	A codon changes but still codes for the same amino acid (GGU → GGC: both glycine)	None
Missense	A codon now codes for a different amino acid	One amino acid swapped; it may or may not matter
Nonsense	A codon becomes a stop codon	The protein is cut short and usually doesn't work
Frameshift	A base is added or deleted	Every codon after it is read wrongly; the protein is scrambled

The best-known missense mutation is the one behind sickle-cell disease. In the haemoglobin gene, the codon GAG (glutamic acid) becomes GTG (valine). One letter in about 400. That single change makes haemoglobin molecules stick together when oxygen is low, which bends red blood cells into a rigid sickle shape that can block small blood vessels.

Frameshifts tend to be worse than swaps, because the problem isn't the one changed base. It's that the ribosome keeps reading in threes from the wrong starting point, so everything downstream turns to nonsense.

The common mix-ups

"DNA goes to the ribosome." It never leaves the nucleus; the mRNA copy does.
"Transcription makes the protein." Transcription makes mRNA. Translation makes the protein.
Forgetting the U. If your mRNA has a T in it, something's gone wrong.
Counting the stop codon as an amino acid. It's a full stop, not a word.

Frequently asked questions

What is a codon?

A codon is a group of three bases on mRNA that stands for one amino acid, or for "stop". AUG codes for methionine and also marks where reading starts; UAA, UAG and UGA are stop codons. With four bases there are 64 possible codons for just 20 amino acids, so most amino acids have more than one codon.

What is the difference between transcription and translation?

Transcription and translation are the two steps of protein synthesis. Transcription happens in the nucleus: RNA polymerase copies one gene from DNA into mRNA. Translation happens at a ribosome in the cytoplasm: the ribosome reads the mRNA one codon at a time while tRNA brings the matching amino acids, and the chain grows into a protein.

Why does the shape of a protein matter?

A protein's job depends on its shape. Enzymes, for example, have an active site that only fits one substrate, which is why changing a single amino acid can stop one working, and why heat that unfolds them is so damaging: see how temperature affects enzyme activity. The ribosomes that build proteins are visible in the Cell Explorer.

Build one yourself

In the DNA Builder you can spin a 3D double helix made from your own sequence, pair the bases yourself, then press Transcribe to watch RNA polymerase build the mRNA, and Translate to watch a ribosome turn it into a chain of amino acids. Click any base to change it, and the builder tells you whether you've made a silent, missense, nonsense or frameshift mutation. Try turning the GAG at codon 7 into GTG: that's sickle-cell, one click away.