Portal

Portal performs two-sample Mendelian randomization analyses using summary statistics from 5,839 GWAS—selecting 401 highest-quality exposure GWAS and the largest published Parkinson's disease (PD) GWAS as outcome—to assess causal relationships between exposures and PD.

Key Features:

• Two-Sample Mendelian Randomization: Integrates summary statistics from 5,839 GWAS to perform two-sample MR analyses between exposures and PD.
• High-Quality Data Integration: Selects the highest-quality exposure GWAS subset (n = 401) and uses summary statistics from the largest published PD GWAS as the disease outcome.
• Causal Effect Assessment: Evaluates causal effects using the inverse variance weighted method and conducts a range of sensitivity analyses to assess robustness.
• Prioritization and False Discovery Rate: Applies a false discovery rate (FDR) of 5% to inverse variance weighted results to prioritize exposures of interest.

Scientific Applications:

• Identification of Causal Exposures: Identified evidence for causal associations between PD and 12 exposures, including nine exposures related to increasing adiposity that associate with decreased PD risk.
• Exploration of Other Risk Factors: Reported nominal associations such as inverse effects of smoking and alcohol consumption on PD risk, and additional noted associations (tea drinking, time spent watching television, forced vital capacity) that were considered potentially biased.

Methodology:

Performs two-sample MR using summary statistics from GWAS (5,839 total; highest-quality exposures n = 401) with the largest published PD GWAS as outcome, estimates causal effects via the inverse variance weighted method with sensitivity analyses, and applies an FDR of 5% to prioritize associations.