A meta-analysis of the results was conducted using the weighted inverse variance method in METAL (http://www.sph.umich.edu/csg/abecasis/metal/index.html)63, which computes a pooled effect estimate (ln(beta)), its standard error and its p-value by weighing the effect estimates of the individual samples by the inverse of it’s variance and by taking into account the direction of effect. Poorly imputed SNPs (r-squared or proper_info < 0.30) and SNPs with low minor allele frequency (MAF<0.01) were excluded, resulting in a final dataset of ~2.4M SNPs. We corrected for any population stratification effects by applying genomic control in each sample prior to meta-analysis. The genomic control inflation factors (λ) for the five personality dimensions for all participating studies ranged between 0.99 and 1.12 (Supplementary Table 1). After applying a genomic control correction to the results from the individual studies, the λ’s for the meta-analyzed results were 1.02, 1.01, 1.03, 1.00 and 1.02, respectively, for Neuroticism, Extraversion, Openness to Experience, Agreeableness and Conscientiousness. The corresponding Quantile-Quantile plots are provided in Supplementary Figure 1. To consider a SNP result genome-wide significant, we used the threshold of P < 5 × 10−8 per trait as proposed for populations of European descent.64
