|
Activity Number:
|
143
- Recent Advances in Imaging Genetics
|
|
Type:
|
Invited
|
|
Date/Time:
|
Tuesday, August 4, 2020 : 10:00 AM to 11:50 AM
|
|
Sponsor:
|
Section on Statistics in Imaging
|
|
Abstract #309265
|
|
|
Title:
|
A Bayesian Spatial Model for Imaging Genetics
|
|
Author(s):
|
Yin Song and Shufei Ge and Jiguo Cao and Liangliang Wang and Nathoo Farouk*
|
|
Companies:
|
University of Victoria and Simon Fraser University and Simon Fraser University and Simon Fraser University and University of Victoria
|
|
Keywords:
|
Imaging Genetics;
Spatial Model;
Variational Bayes;
Conditional Autoregressive Model
|
|
Abstract:
|
We develop a Bayesian bivariate spatial model for multivariate regression analysis applicable to studies examining the influence of genetic variation on brain structure. A bivariate spatial process model is developed to accommodate the correlation structures typically seen in structural brain imaging MRI data. First, we allow for spatial correlation in the imaging phenotypes obtained from neighbouring regions on the same hemisphere of the brain. Second, we allow for correlation in the same phenotypes obtained from different hemispheres (left/right) of the brain. To do this we employ a proper bivariate conditional autoregressive spatial model for the errors in a bivariate spatial regression model. Two approaches are developed for Bayesian computation: (i) a mean-field variational Bayes algorithm and (ii) a Gibbs sampling algorithm. In addition to developing the spatial model and computational procedures to approximate the posterior distribution, we also incorporate Bayesian false discovery rate (FDR) procedures to select SNPs. The methodology is illustrated through a simulation study and an application to data obtained from the Alzheimer's Disease Neuroimaging Initiative study.
|
Authors who are presenting talks have a * after their name.
