ADVANCE Atherosclerotic Disease Vascular
Function and Genetic Epidemiology

Resequencing in the SNP Discovery Set

Shannon and Amita

A sample of 24 men with clinical CAD under the age of 70 years recruited from the Stanford University Medical Center, the Palo Alto VA Hospital, and Kaiser Permanente of Northern California served as our SNP discovery set. DNA was extracted from whole blood and primers were designed to amplify the areas of interest of all the candidate genes by PCR. Purified PCR products were then sequenced using an automated fluorescent labeling system 1, 2 }. Specialized software (Phred, Phrap and Polyphred) was used for lane tracking of the labeled DNA fragments, base calling, sequence assembly, and sequence variant detection. Two highly trained technicians then independently and manually reviewed sequences. Each SNP and genotype was called by hand using the Consed program making sure that base-calls from both strands of DNA agreed. Sequenced SNPs were all cross checked against the public databases.

Our SNP discovery set represents equally four prevalent U.S. race/ethnic groups: Black/African American, East Asian/Asian American, white/European, and Hispanic. For a sample of 48 chromosomes, the detection rate for SNPs with the same minor allele frequency (MAF) across these racial/ethnic groups is >99% for SNPs with a MAF of > 5% and 87% for SNPs with a MAF of > 1% 3. The detection rate for SNPs with significantly different MAF across the four racial/ethnic groups is estimated at 83% for MAF > 5% and 63% for MAF >1% for each ethnic group 3.

Genotyping in the ADVANCE cohorts

Among all SNPs detected by sequencing, we selected a subset to genotype in all participants using the TaqMan® assay 4. We used the ABI prism 7900HT sequence detection system with the Sequence Detection System (SDS2.1) software to make all genotype calls. In general, SNPs with a higher minor allele frequency (to maximize statistical power), SNPs in exonic or promoter regions (to maximize the probability of a functional effect), and SNPs in least linkage disequilibrium (LD) with each other (to identify the maximum number of haplotypes in the region) were selected for genotyping. This selection process was based on visual inspection of all genotypes in the SNP discovery set.

Quality Control

Approximately 100 candidate genes of atherosclerosis were resequenced in our SNP discovery set and genotyped in the ADVANCE cohorts to date. All sequenced SNPs selected for genotyping were first validated in the discovery set with the TaqMan® assay. For the first 505 SNPs genotyped in all ADVANCE cohorts, we were unable to call a genotype in only 0.35% of assays. We also genotyped 15.4% of all samples used for the first 505 SNPs in duplicate. The discrepancy rate among these genotypes was 0.002%.


  1. Strachan T, Read AP. Human molecular genetics 2. New York: Wiley-Liss, 1999:xxiii, 576.
  2. Li J, Tabor HK, Nguyen L, et al. Lack of association between HoxA1 and HoxB1 gene variants and autism in 110 multiplex families. Am J Med Genet 2002; 114:24-30.
  3. Kruglyak L, Nickerson DA. Variation is the spice of life. Nat Genet 2001; 27:234-6.
  4. Li D, Liu L, Chen H, Sawamura T, Mehta JL. LOX-1, an oxidized LDL endothelial receptor, induces CD40/CD40L signaling in human coronary artery endothelial cells. Arterioscler Thromb Vasc Biol 2003; 23:816-21.


Footer Links: