Skip to main content
. Author manuscript; available in PMC: 2015 Dec 1.
Published in final edited form as: Psychophysiology. 2014 Dec;51(12):1309–1320. doi: 10.1111/psyp.12350

Table 2.

Imputation Accuracy into All Available Individuals of European Ancestry

Imputation with MTFS sequences Imputation with 1000 Genomes Aug 2011 release
N Mean RSQ RSQ > .3 N Mean RSQ RSQ > .3
< .001 11,874,879 .479 78.8% 19,028,251 .278 35.5%
MAF [.001–.005) 5,137,513 .639 95.8% 6,641,963 .456 64.9%
MAF [.005–.01) 1,325,092 .764 98.5% 1,690,255 .568 79.7%
MAF [.01–.05) 2,673,865 .862 99.0% 3,022,145 .726 90.1%
MAF ≥ .05 5,872,931 .965 99.5% 6,604,512 .911 97.3%

Note. SNPs were selected using the 1000 Genomes imputed minor allele frequency. RSQ = minimac-estimated quality metric based on the imputation hidden Markov model; RSQ > .3 = percentage of variants with minimac RSQ greater than .3, a conventional threshold used to discard poorly imputed variants.