Fig. 2
From: Leveraging local ancestry to detect gene-gene interactions in genome-wide data
Power comparison for a single causal SNP per locus. Upper panels show the sample size required for 80 % power for the interaction test based on 1 M genotyped GWAS SNPs (S G ) (a), and the interaction test based on local ancestry segment (S L ) assuming a total of 1 K local ancestry segments (b) against the interaction test based on full sequencing data (S S ) assuming a total of 20 M genetic variants (blue curve). Sample size is plotted for increasing ρ GC and ρ GL (defined by the red gradient), the correlation between the true interaction term and the best tag from 1 M genotyped SNPs, and the best tag from local ancestries, respectively. The variance explained by the interaction effect is unrealistically large for illustration purposes and varied between 1 and 10 %. Lower panels show the observed distribution of ρ GC (c) and ρ GL (d) for a randomly selected region from the 1 M Illumina chip and local ancestry, respectively