Genome-wide association filtering using a highly locus-specific transmission/disequilibrium test

Abstract

Multimarker transmission/disequilibrium tests (TDTs) are powerful association and linkage tests used to perform genome-wide filtering in the search for disease susceptibility loci. In contrast to case/control studies, they have a low rate of false positives for population stratification and admixture. However, the length of a region found in association with a disease is usually very large because of linkage disequilibrium (LD). Here, we define a multimarker proportional TDT (mTDT ( P )) designed to improve locus specificity in complex diseases that has good power compared to the most powerful multimarker TDTs. The test is a simple generalization of a multimarker TDT in which haplotype frequencies are used to weight the effect that each haplotype has on the whole measure. Two concepts underlie the features of the metric: the 'common disease, common variant' hypothesis and the decrease in LD with chromosomal distance. Because of this decrease, the frequency of haplotypes in strong LD with common disease variants decreases with increasing distance from the disease susceptibility locus. Thus, our haplotype proportional test has higher locus specificity than common multimarker TDTs that assume a uniform distribution of haplotype probabilities. Because of the common variant hypothesis, risk haplotypes at a given locus are relatively frequent and a metric that weights partial results for each haplotype by its frequency will be as powerful as the most powerful multimarker TDTs. Simulations and real data sets demonstrate that the test has good power compared with the best tests but has remarkably higher locus specificity, so that the association rate decreases at a higher rate with distance from a disease susceptibility or disease protective locus.

Fig. 1

Association rate based on 100 simulations of 200 family trios as a function of the recombination rate using haplotypes of length 4 and different genotype models (rows). The first column shows results for one disease susceptibility locus and the second and third show results for two disease loci. A nominal level of α = 0.05 and a relative risk of 6 were used for all plots. Results for mTDT _P, mTDT and mTDT _Y are plotted in red, blue and green, respectively

Fig. 2

Association rate based on 100 simulations of 200 family trios as a function of the recombination rate using haplotypes of length 10 and different genotype models (rows). The first column shows results for one disease susceptibility locus and the second and third show results for two disease loci. A nominal level of α = 0.05 and a relative risk of 4 were used for all plots. Results for mTDT _P, mTDT and mTDT _Y are plotted in red, blue and green, respectively

Fig. 3

Association rate based on 100 simulations of 200 family trios as a function of the recombination rate using haplotypes of length 10 and different genotype models (rows). The first column shows results for one disease susceptibility locus and the second and third show results for two disease loci. A nominal level of α = 0.05 and a relative risk of 8 were used for all plots. Results for mTDT _P, mTDT and mTDT _Y are plotted in red, blue and green, respectively

Fig. 4

Association rate based on 100 simulations of 500 family trios as a function of the recombination rate using haplotypes of length 4 and different genotype models (rows). The first column shows results for one disease susceptibility locus and the second and third show results for two disease loci. A nominal level of α = 0.05 and a relative risk of 2 were used for all plots. Results for mTDT _P, mTDT and mTDT _Y are plotted in red, blue and green, respectively

Fig. 5

Association rate based on 100 simulations of 500 family trios as a function of the recombination rate using haplotypes of length 10 and different genotype models (rows). The first column shows results for one disease susceptibility locus and the second and third show results for two disease loci. A nominal level of α = 0.05 and a relative risk of 1.6 were used for all plots. Results for mTDT _P, mTDT and mTDT _Y are plotted in red, blue and green, respectively

Fig. 6

Association rate based on 100 simulations of 500 family trios as a function of the recombination rate using haplotypes of length 10 and different genotype models (rows). The first column shows results for one disease susceptibility locus and the second and third show results for two disease loci. A nominal level of α = 0.05 and a relative risk of 2.4 were used for all plots. Results for mTDT _P, mTDT and mTDT _Y are plotted in red, blue and green, respectively

Fig. 7

Association rate based on 100 simulations of 500 family trios as a function of the recombination rate using haplotypes of length 10 and different genotype models (rows). The first column shows results for one disease susceptibility locus and the second and third show results for two disease loci. A nominal level of α = 0.05 and a haplotype length of 10 were used for all plots. Results for mTDT _P, mTDT and mTDT _Y are plotted in red, blue and green, respectively. Simple lines show values for frequency of the disease mutation in the interval [0.2, 0.4] while lines with diamonds show results for mutation frequencies in the interval [0.1, 0.2]

Fig. 8

Comparative TDT maps for loci, a KIAA0350 and b IRF5 data sets using sliding windows of width 6 and offset 1. Rows in gray below each TDT map (colored on a white background) show results for the IHMP data sets as a control test. Results correspond to the following TDTs from top to bottom: mTDT _P (red), mTDT (blue), mTDT _Y (green), mTDT _YP (cyan), mTDT _L1 (purple), mTDT _L2 (scarlet), mTDT _E (violet), mTDT _1T (orange), mTDT _1U (pink), mTDT _LC (black) and mTDT _SR (brown)