Skip to main content
Fig. 4 | BMC Bioinformatics

Fig. 4

From: Identifying novel associations in GWAS by hierarchical Bayesian latent variable detection of differentially misclassified phenotypes

Fig. 4

Improving GWAS discovery using PheLEx in simulations. Simulated true phenotype (green; no misclassification), Simulated misclassified phenotype (blue), and PheLEx corrected phenotype (red). a Box plots for number of true positive SNPs (disease-associated SNPs) (y-axis) found (using Bonferroni-corrected p-value threshold on unadjusted p-values) in association analyses with phenotypes (x-axis): simulated true phenotypes (no misclassification), misclassified phenotypes, and PheLEx corrected phenotypes. b Bar plot for number of simulations (y-axis) where number of novel true positive SNPs (x-axis) were found (using Bonferroni-corrected p-value threshold on unadjusted p-values) in association analyses with PheLEx corrected phenotypes. c Bar plot for number of simulations (y-axis) where number of false positive SNPs (x-axis) were found (using Bonferroni-corrected p-value threshold on unadjusted p-values) in association analyses with simulated true phenotypes (no misclassification), misclassified phenotypes, and PheLEx corrected phenotypes

Back to article page