Fig. 1
From: The hazards of genotype imputation when mapping disease susceptibility variants
Study design schema for the imputation experiments (bottom to top A–E). T2D-associated SNPs were masked simultaneously and imputed using scaffold marker SNPs at A low density and B high density based on all the remaining NGS data, while C masked one T2D-associated SNP at a time. D The full NGS data was phased independently using PHASE v2.1.1, and E the haplotype frequencies in cases and controls were compared with those from the 1000GP haplotype reference panel