The number of SNP–Phenotype tests of association for phenotype-classes varies by PAGE study site genotype and phenotype overlap.

1Number of PAGE study sites where both the SNP and phenotype were available for a given phenotype class. 2Total number of tests of association, by number of PAGE study sites, where both the SNP and phenotype were available for a given phenotype class. 3Number of tests of association that was significant (p<0.01) for two or more PAGE study sites for a single phenotype class and SNP, taking into account matching direction of effect when phenotypically relevant. *Three results where two or more of the groups had a SNP–phenotype association p<0.01 for a single phenotype class across 5 groups represented. The most replicated novel results across studies were for SNPs rs599839, rs10923931, and rs2228145 with hematologic traits.