Based on these simulations, permutation distributions were generated across the chromosomes for each drug-dose phenotype and then used to determine genome-wide LOD score cut-offs corresponding to p-values less than or equal to.05 for each phenotype.
Based on these results we performed 500 000 000 permutations of the entire LD-thinned (r2 = 0.05) height dataset.
Based on these simulations, our expected FDR at p<0.013 is ∼10%, slightly larger than our expected FDR based on permutation of the data.
The LOD threshold was obtained based on permutation test (1000 permutations, P = 0.05) for each data set.
Empirical significance of these findings was estimated based on 1000 permutations through case control randomization within each data set.
In addition, these comparisons were performed using permutation tests based on 5000 permutations.
The permuted P values for the haplotypes are based on 5,000 permutations.
P-values were determined based on permutation analysis using 1,000 permutations.
The possibility of false-positives due to multiple testing is relatively unlikely, because the results of the permutation test [ 17], based on 10,000 random permutations, showed that these associations were significant (data not shown).
