The central dogma

Every living thing runs on instructions written in DNA. The instructions use just four letters: A, C, G and T. A gene is one stretch of that writing, and it tells the cell how to build one particular protein.

The cell does not use the DNA directly. First it copies the gene into a matching molecule called RNA. This copying step is called transcription, like transcribing a page from a locked reference book so you can carry the notes around. The RNA copy is the working version.

Then the cell reads the RNA in chunks of three letters. Each three-letter chunk names one amino acid, and amino acids are the beads that get strung together to make a protein. This reading-and-building step is called translation. Do it bead by bead and a protein grows.

Proteins are the machines and building blocks of you: they digest your food, carry oxygen, build muscle and much more. So the message flows one way, from DNA to RNA to protein, and the protein does the work.

DNA stores the genetic message as a sequence of four bases: adenine, cytosine, guanine and thymine (A, C, G, T). To express a gene, an enzyme called RNA polymerase runs along one strand of the DNA, the template strand, and builds a matching strand of messenger RNA (mRNA). The chemistry is nearly the same, with one swap: RNA uses uracil (U) wherever DNA would have thymine (T). That is transcription.

Translation happens at the ribosome, a molecular machine that reads the mRNA in groups of three bases called codons. Each codon specifies one amino acid. Small adaptor molecules, transfer RNAs (tRNAs), each carry one amino acid and recognise one codon, so the ribosome can add the right amino acid at each step. Reading begins at a start codon, AUG (which also codes for the amino acid methionine), and ends at one of three stop codons, which name no amino acid and release the finished chain.

There are four bases, so there are \(4^3 = 64\) possible codons, but only about 20 amino acids to specify. The genetic code is therefore redundant: several different codons can call for the same amino acid. You can see this in the widget above, where two different codons drop the same bead.

The word "dogma" was Francis Crick's, and it is often misread. His actual claim was narrow and precise: information can flow from nucleic acid to protein, but once it has passed into a protein it cannot flow back out into a nucleic acid sequence. That is different from the cartoon "DNA makes RNA makes protein". Some viruses called retroviruses run RNA back into DNA (reverse transcription), which breaks the simple one-way arrow, but it does not break Crick's real statement, because the information is still moving between nucleic acids, never out of a protein.

What Crick actually said. The central dogma is a statement about the impossibility of one class of information transfer, not a summary of the common pathway. In Crick's formulation, sequence information can pass from nucleic acid to nucleic acid and from nucleic acid to protein, but never from protein back to nucleic acid or from protein to protein. The everyday route DNA → RNA → protein is just the transfer that happens most often; reverse transcription (RNA → DNA) and RNA replication are permitted transfers that Crick explicitly allowed for. The dogma forbids the cell from reading a protein's amino-acid sequence and writing it back into a gene.

Transcription in more detail. RNA polymerase binds at a promoter, unwinds the duplex, and synthesises mRNA in the 5' to 3' direction using the template strand as a guide, so the transcript matches the coding (sense) strand with U for T. In eukaryotes the primary transcript is then processed: a 5' cap and a poly-A tail are added, and introns are removed by splicing, so a single gene can yield several mRNAs (alternative splicing). Only then is the mature mRNA exported for translation.

The code and its structure. The mapping of 64 codons to 20 amino acids plus a stop signal is degenerate, and the degeneracy is not random: synonymous codons usually differ in the third position, which the wobble hypothesis explains through looser base pairing at the third codon base of the tRNA anticodon. Reading frame matters absolutely: inserting or deleting a base that is not a multiple of three shifts the frame and scrambles every codon downstream. Different organisms show codon bias, favouring particular synonymous codons, which affects how fast and how faithfully a message is translated.

Regulation is the real story. Every cell in your body carries the same genome, yet a neuron and a liver cell are utterly different, because not every gene is transcribed and not every transcript is translated. Expression is controlled at many points: which promoters fire, how the mRNA is processed and how long it survives, and whether the ribosome engages. The central dogma tells you the direction information can flow; it says nothing about how much flows, which is where most of biology lives.

Where it gets interesting. mRNA vaccines skip the DNA step entirely: they deliver a ready-made transcript so your own ribosomes translate the target protein directly, training the immune system without a live pathogen. And prions pose a genuine puzzle for the dogma's spirit, because they transmit a trait through protein shape rather than sequence, one misfolded protein forcing its neighbours into the same shape. That is inheritance of a sort, but it carries no new sequence information, so it bends the cartoon without breaking Crick's precise claim.