Table 4.

Association of fast beta EEG variants and DSM-IV AD in subset of discovery sample (N = 2242 from 480 families) and replication sample (N = 1807)

SNP	COGA AA families			Gelernter et al.46 sample
	MAF	Effect size (β)	P-value	MAF	Effect size (β)	P-value
rs11720469	0.451	−0.142	0.070	0.364	−0.154	0.061
rs7428372	0.435	−0.164	0.037	0.364	−0.167	0.042
rs11705903	0.453	−0.161	0.039	0.364	−0.148	0.072
rs9859643	0.449	−0.146	0.060	0.364	−0.149	0.069
rs4680634	0.440	−0.156	0.046	0.364	−0.158	0.055
rs7430210	0.439	−0.157	0.046	0.364	−0.153	0.063
rs7430178	0.440	−0.157	0.046	0.364	−0.165	0.045
rs6806557	0.440	−0.161	0.041	0.364	−0.158	0.057
rs4680631	0.438	−0.148	0.058	0.364	−0.163	0.046
rs13093097	0.460	−0.172	0.027	0.364	−0.179	0.029

Abbreviations: AA, African-American ancestry; AD, alcohol dependence; COGA, Collaborative Study on the Genetics of Alcoholism; DSM-IV, Diagnostic and Statistical Manual of Mental Disorders; EEG, electroencephalography; GWAS, genome-wide association study; MAF, minor allele frequency; SNP, single-nucleotide polymorphism. Note: In COGA AA families, analyses were performed using SAS Version 9.4 (SAS Institute). Models were adjusted for family relatedness, age, sex and PC1–PC10. Individual P-values were adjusted using the P_norm procedure (Wang et al.⁴⁵) that adjusts P-values for multiple testing in the context of both the linkage disequilibrium structure of the SNPs and the dependence structure owing to sampling of relatives. In the sample by Gelernter et al.⁴⁶ (Accession no. phs000425.v1.p1.c1), P-values were extracted from a GWAS of DSM-IV AD (Gelernter et al.⁴⁶). For both samples, individual P-values that withstood the multiple test correction using the P_norm procedure (Wang et al.⁴⁵) are represented in bold.