In silico identification of genetic risk variants for Parkinson’s disease
University of Toronto; King’s College London
My research will shed light on the genetic risk factors for Parkinson’s disease that have not yet been discovered by current methods. A better understanding of the etiology of this and related diseases will help with a more objective diagnosis (in the case of PO), and earlier diagnosis, which are crucial for earlier preventative measures. I will obtain annotations from genomic annotations: Cell level data (Encyclopedia of DNA Elements (ENCODE) project) and tissue-level data (Roadmap Epigenomics Project). Sources for eQTL data include RNA-sequencing experiments conducted by the Genotype Expression Project (GTEx) Project and UKBEC (Ramasamy, 2014). I will then annotate all variants in the 1000 Genomes Project. I will identify which genomic annotations in which tissue are enriched for known PD-associated variants from the NHGRI·EBI Genome-wide Association Study (GWAS) catalogue through tests of enrichment, and also unsupervised statistical learning. The above will then act as inputs into statistical learning algorithms: unsupervised (the algorithm is unaware of which variants are PD-associated; this method is employed to discover any patterns inherent in the data) and supervised (the algorithm is aware of which variants arePD-associated (knowledge derived from the GWAS Catalogue); this method can be employed to develop and test the accuracy of the models derived to predict PD-associated variants).
Neurodegenerative disorders, such as Alzheimer’s disease and Parkinson’s disease, result in a burden on health care systems worldwide. These devastating neurodegenerative diseases are known to be caused in part by genetics, but the genetic aetiology of the illnesses remain largely illusive. Unraveling this mystery will provide the opportunity for earlier and more accurate diagnosis, and the development of treatment options targeting new biological pathways. Numerous studies have demonstrated enrichment of disease-associated variants (for numerous diseases) with genomic functional annotations, modifications to the DNA that affect the accessibility of the DNA to various cellular machinery, and the degree of this enrichment depends on the tissue.
I utilized a recently published method to test for enrichment of various cell types for functional annotations in the genetic component (heritability) of neurodegenerative disorders. I found that variants in genomic annotations from immune cells significantly contribute to the heritability of Alzheimer’s disease. This enrichment suggests that the creation of an accurate model to prioritize neurodegenerative risk variants using these annotations is feasible.