To correct for the large number of SNPs tested per trait, a genome-wide empirical p-value was computed for the asymptotic p-value for each SNP by using 1,000 permutations of swapping sample labels of the traits, using the maxT permutation functionality provided within Plink. A permutation based method using label swapping of the traits is an appropriate method of test correction [18] for these analysis as it is not dependent on these quantitative traits having a normal distribution and also allows the linkage disequilibrium of the genomic regions being tested against the traits to be maintained.
