Tests for association - 2: The transmission disequilibrium test

This is one of a number of tests which aims to be robust against spurious associations due to population stratification by obtaining control alleles from relatives of cases. It is probably better termed the transmission distortion test.

If a parent is heterozygous for a marker, the chances of them transmitting either marker allele to an affected case will be equal unless the marker is linked with the disease gene and unless the marker and disease are associated. A sample of cases and their parents is genotyped and deviations from the expected 50-50 transmission are observed.

Usually the sample consists of a set of trios - affected cases with their parents. However pedigrees containing more than one affected case can also be used.

The A allele is transmitted to affected offspring four times out of five.

If an allele is transmitted to unrelated cases more often than would be expected by chance, this implies that it is linked and associated with the disease mutation. If the sample contains cases related to each other, coming from the same pedigree, then the TDT can become a test only of linkage rather than association.

We have described a method (ETDT - extended TDT) allowing the TDT to be applied to multiallelic markers.