A robust TDT‐type association test under informative parental missingness

Abstract

Many family-based association tests rely on the random transmission of alleles from parents to offspring. Among them, the transmission/disequilibrium test (TDT) may be considered to be the most popular statistical test. The TDT statistic and its variations were proposed to evaluate nonrandom transmission of alleles from parents to the diseased children. However, in family studies, parental genotypes may be missing due to parental death, loss, divorce, or other reasons. Under some missingness conditions, nonrandom transmission of alleles may still occur even when the gene and disease are not associated. As a consequence, the usual TDT-type tests would produce excessive false positive conclusions in association studies. In this paper, we propose a novel TDT-type association test which is not only simple in computation but also robust to the joint effect of population stratification and informative parental missingness. Our test is model-free and allows for different mechanisms of parental missingness across subpopulations. We use a simulation study to compare the performance of the new test with TDT and point out the advantage of the new method.

Detecting Rare Variants in Case-Parents Association Studies

“…Unfortunately, parental genotypes may not be always available. Several methods had been proposed to solve the issue of missing parental genotype, for example, using EM algorithm with the assumption that genotype missing is random conditional on the known genotypes of other family members; see the discussion in Chen and Cheng [37]. These methods are difficult to extend to the cases of multiple variants with very small allele frequencies.…”