Shared polygenic risk and causal inferences in Parkinson's disease

Abstract

Modifiable factors play a key role in modulating the risk of Parkinson's disease (PD). We performed a two-stage genetic analyses to explore genetic correlations and causal relationships between modifiable factors and PD risk. The exposures consisted of GWAS summary statistics of 516 modifiable factors from UK Biobank. The outcome was summary statistics from the large meta-analyses of GWAS on PD. We first utilized linkage disequilibrium score (LDSC) regression to explore genetic correlations between a 516 modifiable factors and PD risk. Two-sample Mendelian randomization (MR) was then used to validate potentially causal relationships between genetically correlated factors and PD risk. LDSC regression showed that PD shared polygenic risk genetic factors with 3 major modifiable factors, including positive correlations with higher educational attainment, and negative correlations with increased adiposity, lower educational attainment and smoking. MR showed causal evidence that increased adiposity was associated with decreased PD risk and higher educational attainment was associated with increased PD risk. Our findings may provide insights in the etiology and pathological mechanisms of PD.

Keywords

• Parkinson's disease
• Modifiable factors
• Linkage disequilibrium score regression
• Mendelian randomization