Skip navigation
Please use this identifier to cite or link to this item:
Title: Context-specific disease gene prediction and functional characterization
Authors: Zhang, Ran
Advisors: Troyanskaya, Olga G
Contributors: Molecular Biology Department
Keywords: complex disease
data integration
machine learning
Subjects: Bioinformatics
Computer science
Issue Date: 2019
Publisher: Princeton, NJ : Princeton University
Abstract: Discovering and understanding disease-associated genes and functional dysregulation are critical for improving human well being. However, the genetic basis and a corresponding functional understanding of disease is still missing, hindered by the noisy, high-dimensional nature of biomedical data, complexity, and heterogeneity of disease mechanisms, as well as the lack of appropriate context-specific human disease samples. To tackle these obstacles, systematic approaches are needed to leverage large-scale biomedical data, functional processes and pathways, as well as context-specific domain knowledge. Here we propose several computational frameworks to investigate disease manifestation in specific cell-types, regions, developmental stages, or other contexts. First, we propose a data-driven framework to study autism spectrum disorder, a complex human neurodevelopmental disorder. We present predictions of autism-associated genes made based on genome-wide, integrative maps of brain pathways, identify autism associated spatiotemporal developmental patterns, functional modules, and the dysregulated processes in autism-associated CNVs. Second, we investigate the role of microglia development in Alzheimer's disease. We construct developmental-stage-specific microglia functional gene interaction networks and discover dysregulation machinery that may be mediating early stages of Alzheimer's disease. Third, we uncover the role of astrocytes in neurological diseases across several brain regions and propose a novel framework to study regional-cell-specific effect when the original data lacks regional and/or disease resolution. Finally, we develop a novel framework for analysis of biomedical literature focused on systematically identifying tissue-specific disease genes. This framework provides a platform to study rare and complex diseases and interpret gene lists from high-throughput biomedical studies (e.g., GWAS, differential expression assays, and drug screens). Taken together, the above work demonstrates the power of developing computational methods that are designed to consider context (such as tissue/cell type, regional, and developmental specificity) to extract biological signal from large compendia of high-throughput data to facilitate understanding of disease.
Alternate format: The Mudd Manuscript Library retains one bound copy of each dissertation. Search for these copies in the library's main catalog:
Type of Material: Academic dissertations (Ph.D.)
Language: en
Appears in Collections:Molecular Biology

Files in This Item:
This content is embargoed until 2021-10-04. For more information contact the Mudd Manuscript Library.

Items in Dataspace are protected by copyright, with all rights reserved, unless otherwise indicated.