RU 2488, Project: Mendelian randomization and polygenic risk scores to understand reduced penetrance in movement disorders

Over the past decades, high-throughput technologies such as next-generation DNA sequencing and genome-wide association (GWA) together with large international consortia have produced an ever-increasing number of phenotype-associated genetic variants have been found for various common complex diseases in humans. To exploit this deluge of data reasonably, stringent dimensionality reduction is required. In this project, we will focus on ways to achieve such enhancement by two state-of-the-art analytical methods, namely Mendelian randomization (MR) and the use of Polygenic Risk Scores (PRS) with the aim to explain the reduced penetrance of known movement disorder mutations.MR is a well-established technique for exploring causality between modifiable risk factors and disease outcome, using genetic variants as instrumental variables with the potential to overcome the problem of confounding and reverse causality. It has become more and more popular over the past years especially given the huge amount of data publicly available from GWA studies. In the first funding period of ProtectMove, we have already shown the promise of this technique for the use in movement disorders, exploring causal factors that facilitate the understanding of penetrance of Parkinson’s Disease (PD). In the second funding period, we will pursue to investigate causal hypotheses for PD, including also PD age of onset, and dystonia with a focus on classical risk factors and inflammation markers (Objective 1), as well as omics markers (Objective 2).PRS allow summarizing the polygenic risk component of a disease in a single numerical value. In part 2 of P8, we will differentiate between PD in mutation carriers (PINK1 (gene encoding phosphatase and tensin homolog-induced putative kinase 1), Parkin, LRRK2 (gene encoding Leucine-rich repeat kinase 2); ‘monogenic PD’) and the more common ‘idiopathic PD’. We will first use simulations to ascertain, under different levels of etiological overlap, how much of the disease risk in monogenic PD can be explained by a PRS for idiopathic PD (Objective 3). After developing new most predictive PRS for both idiopathic and monogenic PD, we will estimate the potential of these scores for modelling the penetrance and variable expressivity of known PD mutations for monogenic PD (Objective 4). Thus, part 2 of P8 will serve to clarify to what extent PRS may facilitate individualized PD risk prediction for mutation carriers of monogenic PD.In the final objective (Objective 5), we will combine MR and PRS. We will use the developed PRS for PD as instrumental variables for MR and investigate the causality of PD for several outcomes such as behavioral traits.The PIs of P8 have extensive expertise in the field of statistical methods for genetic epidemiology. The objectives of P8 will be pursued in close collaboration with P1, P2, P4, P5, P9 and P10 and Core Facility Genetic Replication and Validation in Z2.