SNP-set sequence Kernel Association Test (SKAT) includes an option for efficient resampling of data to obtain empirical p-values. I'm looking for a standard single-locus GWAS tool that similarly has resampling built in.

If there's no efficient resampling built in, I'd have to manually scramble the phenotype data and repeat GWAS with scrambled data many thousands of times. This would be very inefficient.

I wish to resample to correct for non-normality, while avoiding the overcorrection that comes with statistical transformations of data.