Osmanbeyoglu, Hatice Ulku
(2012)
INFORMATION INTEGRATION APPROACHES FOR
INVESTIGATING ESTROGEN RECEPTOR MEDIATED TRANSCRIPTION.
Doctoral Dissertation, University of Pittsburgh.
(Unpublished)
Abstract
Estrogen plays essential roles in the function of normal physiology and diseases. Its effects are mainly mediated through two intracellular estrogen receptors, ERα and ERβ, which belong to a family of nuclear receptors (NRs) functioning as transcription regulators. In the first part of this thesis, we aim to derive a holistic view of the transcription machineries at estrogen-responsive genes and further, to reveal different mechanisms of estrogen-mediated transcription regulation. In order to achieve this, we integrated and systematically dissected a variety of genome-wide high-throughput datasets, including gene expression arrays, ChIP-seq, GRO-seq, and ChIA-PET. Our analyses have led to the following novel findings: In the absence of the ligand, most of the estrogen-responsive genes assumed a high-order chromatin configuration that involved Pol II, ERα and ERα-pioneer factors. Without the ligand, estrogen-induced genes showed active transcription at promoters but failed to elongate into gene bodies, and such a pause was lifted after estrogen treatment. However, the estrogen-repressed genes showed coordinated transcription at promoters and gene bodies in the absence and presence of estrogen. Through information integration, we inferred that, for estrogen-repressed genes, the majority of the high-order chromatin complexes containing actively transcribed genes were disrupted after estrogen treatment. The analyses led to the hypothesis that one mechanism for estrogen-mediated repression is through disrupting the original transcription-favoring chromatin structures.
Further, nuclear receptors such as ERs interact with co-regulators to regulate gene transcription. Understanding the mechanism of action of co-regulator proteins—which do not bind DNA directly, but exert their effects by binding to transcription factors—is important for the study of normal physiology as well as diseased conditions. However, due to the nature of detecting indirect protein-DNA interaction, ChIP-seq signals from co-regulators can be relatively weak and thus biologically meaningful interactions remain difficult to identify. In the second part of this thesis, we investigated and compared different machine learning approaches to integrate multiple types of genomic and transcriptomic information derived from our experiments and from public databases. This helped us to overcome the difficulty of identifying functional DNA binding sites of the co-regulator SRC-1 in the context of estrogen response. Our results indicate that supervised learning with the naïve Bayes algorithm significantly enhanced the peak calling of weak ChIP-seq signals and outperformed other machine learning algorithms. Our integrative approach revealed many potential ERα/SRC-1 DNA binding sites that would otherwise be missed by conventional peak calling algorithms with default settings.
Share
| Citation/Export: |
|
| Social Networking: |
|
Details
| Item Type: |
University of Pittsburgh ETD
|
| Status: |
Unpublished |
| Creators/Authors: |
| Creators | Email | Pitt Username | ORCID  |
|---|
| Osmanbeyoglu, Hatice Ulku | huo1@pitt.edu | HUO1 | |
|
| ETD Committee: |
|
| Date: |
13 December 2012 |
| Date Type: |
Publication |
| Defense Date: |
16 November 2012 |
| Approval Date: |
13 December 2012 |
| Submission Date: |
11 December 2012 |
| Access Restriction: |
5 year -- Restrict access to University of Pittsburgh for a period of 5 years. |
| Number of Pages: |
120 |
| Institution: |
University of Pittsburgh |
| Schools and Programs: |
School of Medicine > Biomedical Informatics |
| Degree: |
PhD - Doctor of Philosophy |
| Thesis Type: |
Doctoral Dissertation |
| Refereed: |
Yes |
| Uncontrolled Keywords: |
gene regulation, bioinformatics, estrogen receptor alpha |
| Date Deposited: |
13 Dec 2012 14:09 |
| Last Modified: |
13 Dec 2017 06:15 |
| URI: |
http://d-scholarship.pitt.edu/id/eprint/16918 |
Metrics
Monthly Views for the past 3 years
Plum Analytics
Actions (login required)
 |
View Item |
![[feed]](/images/feed-icon-32x32.png){kind=icon}