paperKB
coga / coga-kb
Help
Sign in

Chunk #9 — Results — Array design and coverage estimates — Addiction genes

Source
Smokescreen: a targeted genotyping array for addiction research.
Embedded
yes

Text

The array contains 273,494 markers for dense coverage of 1014 genes (±20 kb) across populations. The array captures 97.47, 98.09, 98.08 % of common variants (MAF ≥ 0.05) and 63.61 %, 84.02, and 82.55 % of MAF ≥ 0.01 variants in YRI, ASN, and EUR populations, respectively, directly through linkage disequilibrium (pair-wise r2 ≥ 0.8). The fraction of variants covered by genotype imputation in these gene regions for YRI, ASN, and EUR populations is shown in Fig. 1 (right panel). Consequently, the array has outstanding imputation coverage (98.72, 99.31, and 99.43 %) of common variants and excellent coverage of MAF ≥ 0.01 variants: 85.16, 89.51, and 90.49 % for YRI, ASN, and EUR respectively (obsRSQ ≥ 0.8). In the same order, average obsRSQ of all variants within the addiction gene regions were 0.98, 0.99, and 0.99 for common variants, and 0.91, 0.92, and 0.94 for MAF ≥ 0.01 variants, indicating that imputation works well for most variants in these regions. The list of 1014 genes and each region’s imputation metrics are provided in Additional file 1.