Table 2 Significantly associated loci from GWAS analysis of three phenotypes of interest and increase in the number of hits through the use of imputed phenotypes

From: Deep learning-based phenotype imputation on population-scale biobank data increases genetic discoveries

Phenotype

Missing (%)

AutoComplete (Imp) no. of loci

Observed no. of loci

AutoComplete (ImpAll) no. of loci

More no. of loci

Bilirubin

21

17

42

43

1

LifetimeMDD

80

23

1

29

28

Cannabis ever taken

67

11

1

29

28

  1. GWAS was performed on three phenotypes of interest on the observed, imputed (Imp) and the cohort of all individuals including imputed missing observations (ImpAll). The number of additionally discovered loci (More no. of loci) in applying AutoComplete were tallied in comparison with the original phenotype without imputation.
Back to article page