SNPs included in this analysis were selected from 9 COGA papers reporting family-based association analyses for AD using individuals from the high-density subset (Table 1). The number of individuals included varied across studies: association analyses that encompassed all ancestries ranged from 2139 to 2310 individuals from 262 families; 35 of these families, comprising a total of 298 individuals, are of African American (AA) ancestry. Analyses conducted in the European American (EA) subset ranged from 1172 to 1923 individuals from 217–219 families. Genotyping for these individuals is described in detail in the original COGA papers. Briefly, SNPs within and flanking candidate genes were selected from public databases including dbSNP (http://www.ncbi.nlm.nih.gov/SNP), HapMap (http://www.hapmap.org), and LocusLink (http://www.ncbi.nlm.nih.gov/gene). Genotyping was done using a modified single nucleotide extension reaction, with allele detection by mass spectrometry (Sequenom MassArray system; Sequenom, San Diego, CA, USA). SNPs were in Hardy Weinberg Equilibrium. Genotypes were checked for Mendelian inheritance using programs including PEDCHECK. USERM13 was used to calculate marker allele frequencies and heterozygosities (Edenberg et al., 2008).
