The main figure shows the results of ABPA modeling based on the Sweden +
PGC results (population risk 0.01). The x-axis is the estimated number of SNPs
on a log10 scale, and the y-axis estimates the total variance in
liability explained. The results for five conditions are shown: schizophrenia
(this analysis, red) and, for comparison, results from a published analysis of
myocardial infarction (MI, purple), type 2 diabetes mellitus (T2D, blue), celiac
disease (green), and rheumatoid arthritis (RA, teal). 71 The schizophrenia results are based on
1000 Genomes imputation, and the others on HapMap3 imputation. Color intensity
reflects the probability density with darker colors indicating higher density.
Contour lines show 50% and 95% credible regions for SCZ, and
95% credible regions for the other diseases. The insets depict estimated
SNP distributions for the five disorders: (a) distribution of SNPs in terms of
the variance in liability explained per SNP and (b) the estimated distribution
of SNP genotypic relative risks (GRR). We again stress that multiple qualifiers
are essential in interpreting these estimates.