paperKB
coga / coga-kb
Help
Sign in

Chunk #76 — Methods — Calculating Coverage and Testing for Association

Source
Designing genome-wide association studies: sample size, power, imputation, and the choice of genotyping chip.
Embedded
yes

Text

We used data from the HapMap project Phase II to estimate coverage. Single marker coverage was defined to be the proportion of all variation (with minor allele frequency greater than 5%) in r 2 with a SNP on the genotyping chip above 0.8. Using this definition we achieved very similar estimates to previous studies which used the whole genome (we use twenty two representative megabases). Multi-marker coverage was calculated by an aggressive search of all 2-SNP and 3 SNP haplotypes within 250kb of the SNP being tagged [6]. The SNP was tagged if any of these multi-marker tags had r2 above 0.8, the rule defining the haplotype was also stored and added to the list of multi-marker tests.