coding versus non-coding DNA

Open a thick reference book and you will find that only some pages carry the main entries; the rest is index, cross-references, blank dividers, and instructions on how to use the book. A genome is similar. Only a small slice of DNA spells out protein recipes; the great majority does other things — or, in places, nothing we have yet identified.

Coding DNA is the sequence that is read in triplets (codons) and translated into the amino acids of a protein — the open reading frames within exons. In the human genome this is astonishingly small: only about 1 to 2 percent of the 3.2 billion base pairs actually codes for protein. Everything else is non-coding DNA. That category is not uniform: it includes promoters, enhancers and other regulatory switches; introns; the DNA that is transcribed into functional RNAs that are never translated (ribosomal RNA, transfer RNA, microRNAs, long non-coding RNAs); the structural sequences of centromeres and telomeres; and vast stretches of repetitive and transposon-derived DNA.

The distinction matters because it reshapes how we think about a genome's work. Function is not confined to the coding 1 to 2 percent. A large share of the non-coding genome is regulatory — deciding when and where genes turn on — and disease-linked variants found in genome-wide studies frequently fall in non-coding regulatory DNA rather than in genes. So 'non-coding' means 'does not code for protein', not 'does nothing'. How much of the non-coding genome is genuinely functional versus incidental remains an active, sometimes heated, scientific debate.