|
Abstract:
|
Parkinson's disease is a neurodegenerative disease that primarily affects motor function. Medications currently exist that treat symptoms for a limited time, but there is no cure and symptoms invariably worsen. A better understanding of the genes associated with disease occurrence and progression will be vital to finding a cure. The goal of the work presented in this talk is to identify genes associated with the progression of Parkinson's disease by integrating information from cell biology experiments and pedigree studies. We use a Bayesian hierarchical model to classify genes as being from null, deleterious, or beneficial groups. Based on truncated cell survival data from SiRNA experiments and SNP-based association studies from afflicted family groups, we use MCMC to estimate the posterior probability for group assignment of each gene. By using both information types simultaneously to inform group assignment we can boost power and detect influential genes that may be overlooked using only one data type. This is joint work with biologists at the Gladstone Institutes.
|
