paperKB
coga / coga-kb
Help
Sign in

Chunk #23 — METHODS — Example and validation

Source
SNPinfo: integrating GWAS and candidate gene information into functional SNP selection for genetic association studies.
Embedded
yes

Text

Based on published candidate gene association studies, gene expression studies, and pathway analysis we constructed a list of 848 candidate genes of interest in prostate cancer. Using GenePipe, 542 genes were excluded because none of the GWAS SNPs in these genes had a P-value ≤0.05 and there were sufficient GWAS SNPs to capture (at r2 ≥ 0.8) more than 50% of common (MAF ≥ 0.05) SNPs in Europeans. For the remaining 306 genes, 822 non-redundant SNPs were selected as outlined in Figure 1 with the following GenePipe parameter values: gene upstream region = 5 kb, gene downstream region = 1 kb, MAF = 0.05 for all SNPs, weight = 3 for any predicted functional SNP and small P-value SNPs, weight = 1 for all other SNPs, r2 threshold = 0.8, minimum number of SNPs tagged by each selected tag SNP = 3, minimum number of tag SNPs/gene = 1, and maximum number of tag SNPs/gene = 5.