Skip to main content
. Author manuscript; available in PMC: 2017 Jun 26.
Published in final edited form as: Nat Genet. 2016 Dec 26;49(2):310–316. doi: 10.1038/ng.3751

Table 1.

Interpretation of parameter values in the fitted model. Parameters τ, σ2 and σ3 are dependent on sample sizes, but can be converted to sample-size independent forms (see supplementary note)


Model Interpretation

π1 H0 / H1 Proportion of SNPs not associated with case/control (category 1)
π2 H0 / H1 Proportion of SNPs associated with case/control status but not subgroup status (category 2)
π3 H0 / H1 Proportion of SNPs associated with subgroup status (category 3)
τ H0 / H1 Standard deviation of observed Zd scores (effect sizes for subgroup status) in category 3
σ2 H0 / H1 Standard deviation of observed Za scores (effect sizes for case/control status) in category 2
σ3 H1 only Standard deviation of observed Za scores (effect sizes for case/control status) in category 3
ρ H1 only ‘Absolute covariance’ between Zd scores (effect sizes for subgroup status) and Za scores (effect sizes for case/control status) in category 3