Rare genetic variants in Parkinson's Disease: an evolutionary and high-throughput resequencing approach

Recent genome-wide association studies (GWAS) related to Parkinson Disease (PD) have identified a limited number of genetic loci reaching genome-wide significance. Now that our scientific field is entering the post-GWAS era, it is essential that these GWA studies are followed up systematically since (i) a number of GWAS signals fail to be replicated consistently, (ii) the majority of these GWAS signals have not been tracked to clear causal variants, and (iii) additional genes known to be mutated in Mendelian forms of parkinsonism do show association evidence in GWAS. Our initial hypothesis is that a resequencing approach based on next-generation technologies can describe the whole genetic variation associated with PD and therefore be informative on sources of genetic variation poorly captured by existing GWAS approaches. For that, we propose to resequence all coding regions and known regulatory elements of the 41 genes most clearly related with PD as detected by GWAS and/or that cause Mendelian forms of the disease on DNA samples already available from 240 PD patients and 186 unrelated controls. Since we will be exploring the complete genetic variation spectrum of this set of candidate genes in PD cases, we expect to be able to pinpoint which functional variants (rare and/or common) underlie the causality of the disease. Our secondary hypothesis is that an excess of rare variants can signal the involvement of a gene in a complex disease such as PD, and that an evolutionary approach can test whether such an excess is present. We propose to evaluate the role of rare variants to PD by making use of neutrality statistics such as Tajima's D, in which deviations of the allele frequency spectrum towards excess of rare variants can be seen even with moderate sample sizes. Moreover, these tests can be applied to single genes, their parts, or to several concatenated genes, in order to focus on specific gene regions (such as regulatory regions) or to increase their statistical power. Finally, different in silico approaches will be applied to determine the possible functional consequences of both coding and non-coding variants found associated with cases in order to restrict the number of those to experimentally test by means of functional or expression assays. Thus, the overall expected deliverable of this project is a much clearer understanding of the implication of this set of genes in the etiopathogenesis of PD.