Additional file 1 of Genetic proxies for antihypertensive drugs and mental disorders: Mendelian randomization study in European and East Asian populations

Additional file 1: Supplementary Methods. Flow chart of instrument selection and study design. Figure S1. Associations of BBs and CCBs using SNPs selected from LD threshold at r2< 0.01 in the European population. Two SNPs were used for genetic proxies for BBs and 8 SNPs were used for genetic proxies for CCBs. Figure S2. MR analysis for CCBs excluding pleiotropic SNPs related to CACNA1C gene in the European population and excluding pleiotropic SNPs related to CACNAB2 gene in the East Asian population. 22 SNPs served as genetic instruments in the European population and 14 SNPs served as genetic instruments in the East Asian population. Figure S3. MR analysis for CCBs excluding pleiotropic SNPs related to BMI. After removing rs3821843 and rs10828399 for Europeans and rs61842677 for East Asians related to BMI, 22SNPs served as genetic instruments in the European population and 20 SNPs served as genetic instruments in the East Asian population. Figure S4. MR sensitivity analysis for CCBs excluding pleiotropic SNPs related to cause of death. After removing two SNPs (rs113210396 and rs72786098) related to cause of death for the European population, 22SNPs served as genetic instruments for CCBs. Figure S5. MR sensitivity analysis for CCBs excluding pleiotropic SNPs related to BMI, CACNA1C gene, and cause of death. After removing all six SNPs related to CACNA1C, BMI and cause of death for the European population, 18 SNPs served as genetic instruments for CCBs. Figure S6. Associations of CCBs using SNPs selected from LD threshold r2< 0.01 and excluded pleiotropic SNPs related to BMI and CACNA1C gene. After removing 2 SNPs related to risk genes and potential confounder BMI for the European population, 6 SNPs served as genetic instruments. Figure S7. MR analysis of gene-specific effects of CCBs. Nsnp (column 4) indicates the number of SNPs aggregated for each gene targeted region of CCBs. Figure S8. Visual representation of the MR-Egger estimates of the genetic associations of CCBs with BD and SCZ. Figure S9. Bayesian colocalization test of ACEIs with schizophrenia. Panel A: The posterior probability for the distinct causal variants model (PPH3) is 0.011. The posterior probability for the model with a shared causal variant (PPH4) is 0.98. Panel B: The posterior probability for the distinct causal variants model (PPH3) is 0.054. The posterior probability for the model with a shared causal variant (PPH4) is 0.35. Table S1. Information on genetic instruments. Table S2. Power calculation. power calculation for detectable odds ratio (per 5 mmHg lower in SBP) at 80% power. The R2 was calculated as beta2* 2 * MAF * (1-MAF), where beta is the SNP-outcome association standardized to the phenotypic variance and MAF is the minor allele frequency of the SNP. The R2 for ACEI among Europeans is 0.0015, and among the East Asians is 0.0016. The R2 for BB among the Europeans is 0.009, and among east Asians is 0.0018. The R2 for CCB among the Europeans is 0.030, and 0.046 among the East Asians. Table S3. Sensitivity analysis using eQTL SNPs to proxy antihypertensive drugs in the European population. Table S4. I2GX of CCBs MR-Egger analysis. Table S5. Sensitivity analysis using GWAS of SBP in the UK Biobank without adjustment of BMI. Table S6. Sensitivity analysis for bipolar disorder GWAS without UK Biobank participants. Table S7. Sensitivity analysis using DBP-associated SNPs to proxy antihypertensive drugs in the European population.