Such blocks tend to vary in length in the human population. Such variation is thought to arise during replication of the DNA, when the enzyme that copies DNA, DNA polymerase, slips and accidentally inserts an extra CAG, or misses one out. This slippage is fairly rare, however, so although there is variation between individuals, the blocks are inherited stably from generation to generation.

Sequence repeats in genes can occasionally lead to human disease, such as, for example, Huntington's disease, a neurodegenerative disease that occurs in about 5 per 100 000 people.

Thousands of other sequence repeats, with no biological consequences, have been mapped in the genome. As well as being used for 'fingerprinting' - in paternity testing, forensics and so on - these repeats have been used for many years as 'markers' that can identify specific points in the genome.

Researchers can track markers through family pedigrees to find those associated with a biological characteristic, typically an inherited disease. As the markers act as 'landmarks' in the genome, researchers can then look for candidate genes nearby.

Many human disease genes have been identified this way, although maps based on repeat markers are not ideal as they are 'low resolution' - with millions of nucleotides between the markers - making it difficult to pinpoint a disease-related region of DNA.

Image credit: Kate Whitley