Supplementary Material for: Polymorphisms and Noncardioembolic Stroke in Three Case-Control Studies

<i>Background:</i> Gene variants associated with disease could reveal novel mechanisms. We searched for single nucleotide polymorphisms (SNPs) associated with noncardioembolic stroke (nonCES). <i>Methods:</i> We tested 24,926 SNPs in or near genes for association with nonCES in the Vienna Study (551 cases, 815 controls) and then evaluated the associated SNPs in the UCSF-CC Study (570 cases, 1,604 controls) first in pooled DNA samples and then in individual DNA samples. We then asked whether the risk alleles of the SNPs associated with increased risk in both studies were also associated with increased risk of nonCES in the German Study (728 cases, 1,041 controls). <i>Results:</i> Six of the 46 SNPs that were associated with nonCES in both the Vienna and the UCSF-CC Studies were also associated with nonCES in the German Study: rs362277 in <i>HTT</i> (OR 1.39, 90% CI 1.12–1.71), rs2924914 near <i>CSMD1</i> (OR 1.22, 90% CI 1.04–1.43), rs1264352 near <i>DDR1</i> (OR 1.20, 90% CI 1.02–1.41), rs544115 in <i>NEU3</i> (OR 1.63, 90% CI 1.02–2.62), rs12481805 in <i>UMODL1</i> (OR 1.31, 90% CI 1.01–1.81), and rs2857595 near <i>NCR3</i> (OR 1.15, 90% CI 1.00–1.32). Accounting for multiple testing of 46 SNPs, these 6 SNPs had a false discovery rate of 0.69. <i>Conclusions:</i> Some of the 6 SNPs may be associated with nonCES but most may be false positives. These 6 SNPs merit investigation in additional nonCES study populations.