Transmission/disequilibrium test when neither parent is available in some families: a non-iterative approach.

Abstract

The transmission/disequilibrium test (TDT) is a simple method of detecting linkage between a marker locus and a disease-susceptibility locus. The test requires genotype data from the affected individuals (the probands) and their parents, the 'case-parents triads'. However, missing data from parents often complicate the situations. Several non-iterative methods, such as the sibling-based TDT, the reconstruction-combined TDT, and the 1-TDT have been proposed to deal with the problem. But when the father, the mother, and other non-proband siblings in a family are all unavailable for study, that family has to be excluded from analysis by any of the above methods. The author proposes a new non-iterative method to deal with the situation when the data contain some 'orphaned' probands. The method utilises the complete data from case-parents triads to infer probabilistically the missing parental genotypes, under the assumption that the missingness is independent of genotype and ethnic origin. The methods can be tailored to specific genetic models, such as the gene-dose, dominance, and recessive alternatives. Simulation analysis shows that, under the null hypothesis of no linkage, the proposed method is not prone to produce an excess of false positive results, even in complex situations such as population stratification or assortive mating. Under the linkage alternative, the new method recaptures much of the information that would be lost by discarding the orphaned probands. The method provides a way for dealing with the situation where neither parents nor siblings are available in a TDT study.