Significant evidence for linkage of a simulated trait to D1G024--a conclusion reached using multiallelic...

We applied three versions of the transmission/disequilibrium test (TDT) for detecting linkage in nuclear families using the a priori information of possibly present association. For our analyses we employed all marker data for a simulated trait. Results for replicate 11 showed significant linkage to D1G024 using the multiallelic Tmhet statistic and the extended TDT. The TDT for the most frequent parental allele which had been applied successfully in previous studies failed to detect linkage. This result may have happened because the data did not arise from a realistic simulation of the evolution of a real population, where mutations might be expected to occur against a specific haplotype. Simulation studies are required to elucidate the applicability of the TDT for the most frequent parental allele in realistic situations. Furthermore, guidelines have to be developed for how to pool categories in large, sparse contingency tables to obtain larger cell frequencies and lower degrees of freedom.