Predicting risk of Alzheimer’s disease using polygenic risk scores developed for Parkinson’s disease

Abstract

AbstractBackground and ObjectivesThe two most common neurodegenerative diseases are Alzheimer’s disease (AD) and Parkinson’s disease (PD), both related to age and affect millions of people across the world, especially as life expectancy increases in certain countries. Here, we explore the potential predictiveness of the genetic risk of AD and PD separately and then the extent of the underlying shared genetics of AD and PD.MethodsThe population genetic risk estimates for AD and PD were derived using a previously developed population specific polygenic risk score (psPRS) and regression-based SNP filtering method. To test the overlap between AD and PD, we ran a regression of the AD psPRSs versus the population PD prevalences for both the filtered and unfiltered AD PRS and vice versa. We then assessed gene-gene interaction and pathway involvement using the Alzheimer’s KnowledgeBase (AlzKB) and STRINGdb, respectively.ResultsThe unfiltered PD psPRS was moderately predictive, while the AD psPRS was not. After filtering, both the AD and PD, psPRSs improved to strongly predictive, explained most of the genetic variation. The ability of the unfiltered AD psPRS to predict PD, and vice versa, is poor. However, the filtered AD, and PD, psPRS were highly predictive.DiscussionOur results suggest that there is a correlation between AD, and PD, specific allele frequency and prevalence, as well as an overlap of AD and PD generally. However, the AD psPRS is a better predictor of PD, than the PD psPRS is of AD. Our results call for further research into the general overlap of Alzheimer’s disease and Parkinson’s disease, despite the previous lack of evidence.