G-WIZ

G-WIZ computes receiver-operator characteristic (ROC) curves and area under the ROC (AUROC) values from summary-level GWAS data to quantify the predictive performance of SNP-based disease risk models.

Key Features:

• AUROC Calculation: Calculates receiver-operator characteristic (ROC) curves and area under the ROC (AUROC) values to quantify predictive accuracy of SNP biomarkers.
• Validation: Validated to predict AUROC values with less than 3% error when compared against patient-level SNP data and literature-reported AUROC values.
• Summary-Level Data Utilization: Operates on summary-level GWAS data rather than individual-level genotype data.
• Database Integration: Utilizes summary-level GWAS data from GWAS Central and has been applied to 569 GWA studies covering 219 distinct medical conditions.
• Explained Heritability Metrics: Computes explained heritability metrics from summary-level GWAS data.
• Performance Insights: Provides comparative AUROC analyses reporting an average AUROC of approximately 0.55 across studies and individual studies with AUROC values above 0.75.

Scientific Applications:

• GWAS predictive performance assessment: Evaluating the accuracy of SNP-based disease risk predictions derived from genome-wide association studies.
• Genetic epidemiology: Quantifying the contribution of SNPs to disease prediction for genetic epidemiology analyses.
• Biomarker evaluation for personalized medicine: Assessing the potential and limitations of GWAS-derived biomarkers for personalized medicine and clinical decision-making.

Methodology:

Implemented as an R package that generates ROC curves and computes AUROC and explained heritability metrics from summary-level GWAS data (sourced from GWAS Central), with validation by comparison to patient-level SNP data and literature-reported AUROC values.