Chunk #22 — Using Results from the GWAS of the International Schizophrenia Consortium as an Example, Are the Results of Dickson et al. Supported by Empirical Observation?
In our simulations, as in those of Dickson et al., we observed genome-wide significant results a good proportion of the time (90% for loci with 9 causal variants with GRR = 4), suggesting that if variants of this effect size and frequency ∼0.01 exist (the benchmark simulations of Dickson et al.) then they would have been detected by standard single SNP analysis in the ISC GWAS. It is important to note that this ability is achieved using standard GWAS chips, which are designed to have a biased distribution of allele frequencies (higher proportion of common variants) relative to the distribution of SNP allele frequencies in the population (and the distribution of SNP frequencies generated by the coalescent simulations).
