Home Page 1

Reference URL

Comment

Save to favorites

To link to the entire object, paste this link in email, IM or document

To embed the entire object, paste this HTML in website

To link to this page, paste this link in email, IM or document

To embed this page, paste this HTML in website

Systematic identification of potential therapeutic targets in cancers using heterogeneous data

Page 1

Save page

1 of 161

View Description

View PDF & Text

Download

small (250x250 max)

medium (500x500 max)

large ( > 500x500)

Full Resolution

All (PDF)

This page

All

Subset

PDF
Text

Text Search...

Object Description

Title	Systematic identification of potential therapeutic targets in cancers using heterogeneous data
Author	Wu, Chia-Chin
Author email	chiachin@usc.edu; thelmaliao@hotmail.com
Degree	Doctor of Philosophy
Document type	Dissertation
Degree program	Biomedical Engineering
School	Viterbi School of Engineering
Date defended/completed	2010-06-17
Date submitted	2010
Restricted until	Unrestricted
Date published	2010-08-05
Advisor (committee chair)	D'Argenio, David
Advisor (committee member)	Triche, Timothy Khoo, Michael Asgharzadeh, Shahab
Abstract	A major goal of cancer research is to find specific therapies to molecular targets in cancers. Recently, great amounts of data from a variety of high-throughput technologies and public in silico resources offer the possibility of understanding cellular mechanisms and aiding drug discovery process. This demonstrates a clear need for developing computational strategies to comprehensively and speedily search effective therapeutic targets in cancers using integrated biological resources. Therefore, this research proposes to take advantage of these resources to develop a genetic network-based approach to comprehensively and effectively identify potential therapeutic targets in cancers.; First, multiple genomic and proteomic datasets are integrated to construct a human genetic network that is able to reveal the tendency for genes to operate in same pathways. The research here addresses three major problems that confront the construction of a human genetic network from heterogeneous genomics data at the whole-genome level: definition of a robust Gold Standard Negative (GSN) set, large-scale learning, and massive missing data values. A graph-based approach was proposed to generate a robust GSN for the training process of genetic network construction. The developed Relevance Vector Machine (RVM)-based ensemble model that combines AdaBoost and Reduced-Feature yields improved performance on large-scale learning problems with massive missing values in comparison to Naïve Bayes, the most popular approach used to construct genetic and protein networks.; Second, an approach is developed to discover cancer-related genes or potential therapeutic targets by taking advantages of the constructed genetic network. A tumor-specific network is generated by mapping the differentially expressed genes in a cancer to the constructed whole-genome genetic network. The differentially expressed genes are then all ranked based on the extent of their functional association with multiple known cancer pathways in the tumor-specific network. Those highly ranked genes are considered as potential therapeutic targets in the cancer. The approach has been applied to several cancer types: Breast, Colon, and Lung Cancer separately. The result in each case shows that higher ranked genes are cited by more literature respectively related to the three cancers; that is, they likely play more important roles in these cancers, compared to lower ranked genes. While mapping the results to gene functional annotation, we find that many kinase, receptor, and transcription factor related genes, which are often proposed as possible molecular targets, are ranked highly in all cases. The target genes detected by siRNA screens also tend to be ranked highly in the prediction. Additionally, by mapping the results to drug-target information, most targets of drugs, already in clinical trials and used in the treatments of the three cancers, are found to be highly ranked in each case. Other drugs and compounds whose targets are also highly ranked have also shown anti-cancer effect and could be considered as potential novel drugs for these cancers. Moreover, we find some highly ranked targets in each case not yet identified as cancer genes. Taken together, the results presented in this work suggest that the genetic network-based approach on the basis of a systems-biology perspective will play a significant and increasing role in cancer drug discovery in the future.
Keyword	cancer therapeutic targets; cancer systems biology; genetic network; relevance vector machine (RVM), ensemble learning
Language	English
Part of collection	University of Southern California dissertations and theses
Publisher (of the original version)	University of Southern California
Place of publication (of the original version)	Los Angeles, California
Publisher (of the digital version)	University of Southern California. Libraries
Provenance	Electronically uploaded by the author
Type	texts
Legacy record ID	usctheses-m3296
Rights	Wu, Chia-Chin
Repository name	Libraries, University of Southern California
Repository address	Los Angeles, California
Repository email	http://www.usc.edu/isd/libraries/services/ask_a_librarian/email/
Filename	etd-Wu-3970
Archival file	uscthesesreloadpub_Volume35/etd-Wu-3970.pdf