In the primary GWAS sample, individuals with elevated fast beta EEG were more likely to meet criteria for DSM-IV AD, CoD and ADHD (all P-values <0.001, surviving Bonferroni’s correction). To determine whether the significant signal observed for fast beta EEG was accounted for by one of these disorders, we carried out three separate post hoc GWAS of fast beta EEG, each with one of these phenotypes included as a covariate in the model. Including AD as a covariate, the 3q26 association remains but is slightly diminished (rs11720469: β: − 0.120, P-value: 2.2 × 10−8; Supplementary Figure 3A). Including DSM-IV CoD as a covariate, the 3q26 association remains but is also slightly diminished (rs11720469: β: − 0.122, P-value: 1.3 × 10−8; Supplementary Figure 3B). Including DSM-IV ADHD as a covariate reduced the 3q26 association (rs11720469: β: − 0.088, P-value: 6 × 10−4; Supplementary Figure 3C).
