Linkage analysis is the technique typically used to determine the genetic location of a disease gene when there are no other 'signposts' available (no cytogenetic abnormality, co-inherited disorders, good candidate genes, or known protein product). The goal of linkage analysis is to identify a piece of DNA of known location that is inherited by all family members affected by the disorder being studied, and is not inherited by any of the unaffected family members. This piece of DNA is said to be 'co-inherited' or to 'co-segregate' with the disease phenotype. Once this piece of DNA is found, one knows that the disease gene must lie somewhere close by. Determining the location of the disease gene is the first step toward identifying the gene itself.

The pieces of DNA of known location that are used in linkage analysis are called 'polymorphisms'. Polymorphisms are benign differences in DNA found among humans that are not related to any clinical conditions (they are not disease mutations). Polymorphisms are abundant in the human genome, and can be found between and close to virtually all genes. By identifying a co-inherited polymorphism, testing additional polymorphisms in the vicinity, and then using statistical programs to calculate the odds that you are correct in the choice of location (the LOD score), the disease gene can be placed within a 'critical region' flanked by key polymorphic markers and measured using a genetic distance referred to as a cM (centiMorgan).

The most common type of polymorphism currently used for linkage analysis is called a 'short tandem repeat' (STR). Common STRs are dinucleotide repeats, in which the nucleotides C and A are repeated ~15-30 times in a row (CACACACACACACACACACACA...), and tetranucleotide repeats, in which four base pairs are repeated (GATAGATAGATAGATAGATAGATA...). These repeats occur at the same place in each of our genomes, but the number of times each polymorphism is repeated in a given person's genome varies between individuals. Because a given STR only changes in size once every 100 generations or so, its size can be measured in each family member and its inheritance can be traced through the family.

Many thousands of STRs have been identified and mapped to specific locations on each chromosome. Primers are made to the unique DNA sequence to each side of a given repeat, and these primers are used to amplify the repeat using the polymerase chain reaction (PCR). The millions of copies of the repeat are either radioactively or fluorescently labeled and then run on a gel to separate the different sizes from one another. The size of each sequence, which correlates with the number of repetitive sequences within it, can then be assessed.