Wang, Xingbin
(2012)
Genomic Meta-Analysis Combining Microarray Studies with Confounding Clinical Variables: Application to Depression Analysis.
Doctoral Dissertation, University of Pittsburgh.
(Unpublished)
Abstract
Major depressive disorder (MDD) is a heterogeneous psychiatric illness with mostly un-characterized pathology and is the fourth most common cause of disability according to the World Health Organization (WHO) and has a significant impact on public health in the United States. To understand the genetics of MDD, we aim to develop effective meta-analysis approaches to provide high-quality characterization of MDD related biomarkers and pathways with proper clinical variable adjustment. First, genomic meta-analysis in MDD faces multiple unique difficulties, such as weak expression signal of MDD, substantial clinical heterogeneity and small sample size. Given these obstacles, it is hard to identify consistent and robust biomarkers in an individual study. To achieve a more accurate and robust detection of differentially expressed (DE) genes and pathways associated with MDD, we proposed a statistical framework of meta-analysis for adjusting confounding variables (MetaACV). The result showed that more MDD related biomarkers and pathways were detected that greatly enhanced understanding of MDD neurobiology. Secondly, Meta-analysis has become popular in the biomedical research because it generally can increase statistical power and provide validated conclusions. However, its result is often biased due to the heterogeneity. Meta-regression has been a useful tool for exploring the source of heterogeneity among studies in a meta-analysis. In this dissertation, we will explore the use of meta-regression in microarray meta-analysis. To account for heterogeneities introduced by study-specific features such as sex, brain region and array platform in the meta-analysis of major depressive disorder (MDD) microarray studies, we extended the random effects model (REM) for genomic meta-regression, combining eight MDD microarray studies. The result shows increased statistical power to detect gender-dependent and brain-region-dependent biomarkers that traditional meta-analysis methods cannot detect. The identified gender-dependent markers have provided new biological insights as to why females are more susceptible to MDD and the result may lead to novel therapeutic targets. Finally, we present an open-source R package called Meta-analysis for Differential Expression analysis (MetaDE) which provides 12 commonly used methods of meta-analysis. It is a friendly used software such that biologists implement meta-analysis in their research.
Share
| Citation/Export: |
|
| Social Networking: |
|
Details
| Item Type: |
University of Pittsburgh ETD
|
| Status: |
Unpublished |
| Creators/Authors: |
|
| ETD Committee: |
|
| Date: |
30 January 2012 |
| Date Type: |
Completion |
| Defense Date: |
15 August 2011 |
| Approval Date: |
30 January 2012 |
| Submission Date: |
29 November 2011 |
| Access Restriction: |
2 year -- Restrict access to University of Pittsburgh for a period of 2 years. |
| Number of Pages: |
124 |
| Institution: |
University of Pittsburgh |
| Schools and Programs: |
School of Public Health > Biostatistics |
| Degree: |
PhD - Doctor of Philosophy |
| Thesis Type: |
Doctoral Dissertation |
| Refereed: |
Yes |
| Uncontrolled Keywords: |
MDD, meta-analysis,microarray,meta-regression |
| Date Deposited: |
30 Jan 2012 19:11 |
| Last Modified: |
15 Nov 2016 13:55 |
| URI: |
http://d-scholarship.pitt.edu/id/eprint/10570 |
Metrics
Monthly Views for the past 3 years
Plum Analytics
Actions (login required)
 |
View Item |
![[feed]](/images/feed-icon-32x32.png){kind=icon}