Using Omics Data to Guide Network Classification in Neuroimaging Studies of Brain Diseases

• Activity Number: 562 - Integrating Neuroimaging and Genomics Data
• Type: Invited
• Date/Time: Wednesday, August 1, 2018 : 2:00 PM to 3:50 PM
• Sponsor: Section on Statistics in Imaging
• Author(s): Jean Yee Hwa Yang* and Elizaveta Levina and Mengbo Li and Jesús Arroyo and Daniel A. Kessler
• Companies: University of Sydney, Australia and University of Michigan and University of Sydney and University of Michigan and University of Michigan
• Keywords: transcriptomics; network; brain connectome

Abstract

Gene expression studies are playing an increasingly important role in understanding human brain functions and will ultimately contribute to the development of therapies and treatments to neurological diseases (Sunkin et al., 2013). The Allen Brain Atlas is a collection of data sets providing extensive gene expression, connectivity and neuroanatomical data, including spatially mapped microarray data from human brain. In this talk, we will present an integrative approach for multi-model data from spatial prospective that utilize the imaging and omics data from publically available brain researches. We will describe the construction of informative weights or measures to guide network classification in high-dimensional data with a special focus to enable meaningful interpretation of feature selection. Instead of focusing on matching sample information in multi-model integration. Finally, we will present the results that examines and evaluation of various informative weights from omics data to guide image classification.