Statistical aspects of the transmission/disequilibrium test (TDT)

摘要

The transmission/disequilibrium test (TDT) was introduced as a direct test of linkage which is not affected by the problem of population stratification. Such a test is needed since much of the data used currently for linkage tests does come, or might be suspected to come, from stratified populations. Also, the test is valid when the data include relatives, since it overcomes the dependence properties usually associated with such data. In practice, the main purpose of the procedure is to test for linkage between a marker locus and a purported disease locus - a link in the chain of activities whose ultimate aim is to locate disease loci. The test differs from frequently used tests based on sharing of marker alleles between affected relatives and, unlike sharing tests, is related to the population concept of association. These differences are discussed. Many interesting questions arise in the statistical theory of the TDT, some of which are still unresolved. One of the aims of this paper is to raise and discuss these.