Selected variants in top admixture mapping association regions, and their association testing results in stratified BP GWAS (stratified by genetic analysis groups and followed by a step of meta-analysis across the groups).

All admixture mapping hits here were in the analysis of Amerindian ancestry versus other ancestries. We provide the effect estimate (Effect) from both the admixture mapping (Ancestry effect), and association mapping, the frequency of the ancestry (Ancestry freq), and the frequencies of the variant in the Mexican (high Amerindian ancestry) and Cuban (low Amerindian ancestry) genetic analysis groups. Genomic positions are in human build 37. A1 is the effect (tested) allele, A2 is the other allele. Type refers to imputation: ‘g’ is genotyped, ‘i’ imputed. Info is a measure of imputation quality. a For SNP rs118163160, the association analysis results when restricted to the Mainland groups were effect = −2.85, SE = 0.69, p-value = 3.84E-05. P-value for heterogeneity of the effect across the genetic analysis groups was 0.001.