A meta-analysis of the GWA results of the discovery cohorts was conducted using the weighted inverse variance method in METAL (http://www.sph.umich.edu/csg/abecasis/metal/index.html). This is a fixed effects model in which effect sizes (beta’s) are weighted by the inverse of their variance and then summed over cohorts. This model is appropriate if phenotypes are on a similar scale, which was the case for the harmonized neuroticism scores. Poorly imputed SNPs (r2<0.30 or proper_info<0.40) and SNPs with low MAF (MAF < √(5/N, which corresponds to less than 5 estimated individuals in the least frequent genotype group, under the assumption of HWE) were excluded, resulting in a total number of 1.1M to 6.6M SNPs across cohorts. The number of unique SNPs available for meta-analysis was 7,480,565. For 530,951 SNPs association results were available in one cohort only and were discarded, leading to a final 6,949,614 SNPs for which results are reported. Genomic control inflation factors (lambda) and Manhattan and quantile–quantile plots per cohort are provided in eTable 3 and eFigures 1 and 2. SNPs with a P-value of 5 × 10−8 or smaller were
