Files and links (1)
PDFOpen Access (License Unspecified), Open Access
Dissertation
Systems toxicology: mining chemical-toxicity signaling paths to enable network medicine
Doctor of Philosophy (Ph.D.), Drexel University
May 2012
DOI:
https://doi.org/10.17918/etd-7456
Abstract
Systems toxicology, a branch of toxicology that studies chemical effects on biological systems, presents exciting knowledge discovery challenges for the information researcher. The exponential increase in availability of genomic and proteomic data in this domain needs to be matched with increasingly sophisticated network analysis approaches. Improved ability to mine complex gene and protein interaction networks may eventually lead to discovery of drugs that target biological sub-networks ('network medicine') instead of individual proteins. In this thesis, we have proposed and investigated the use of a maximal edge centrality criterion to discover drug-toxicity signaling paths inside a human protein interaction network. The signaling path detection approach utilizes drug and toxicity information along with two novel edge weighting measures, one based on edge centrality for detected paths and another using differential gene expression between tissues treated with toxicity-inducing drugs and a control set. Drugs known to induce non-immune Neutropenia were analyzed as a test case and common path proteins on discovered signaling paths were evaluated for toxicological significance. In addition to investigating the value of topological connectivity for identification of toxicity biomarkers, the gene expression-based measure led to identification of a proposed biomarker panel for screening new drug candidates. Comparative evaluation of findings from the DTSP approach with standard microarray analysis method showed clear improvements in various performance measures including true positive rate, positive predictive value, negative predictive value and overall accuracy. Comparison of non-immune Neutropenia signaling paths with those discovered for a control set showed increased transcript-level activation of discovered signaling paths for toxicity-inducing drugs. We have demonstrated the scientific value from a systems-based approach for identifying toxicity-related proteins inside complex biological networks. The algorithm should be useful for biomarker identification for any toxicity assuming availability of relevant drug and drug-induced toxicity information.
Metrics
28 File views/ downloads
20 Record Views
Details
- Title
- Systems toxicology
- Creators
- Kaushal D. Desai - DU
- Contributors
- Xiaohua Hu (Advisor) - Drexel University (1970-)
- Awarding Institution
- Drexel University
- Degree Awarded
- Doctor of Philosophy (Ph.D.)
- Publisher
- Drexel University; Philadelphia, Pennsylvania
- Number of pages
- xi, 97 pages
- Resource Type
- Dissertation
- Language
- English
- Academic Unit
- College of Information Science and Technology (1995-2013); Drexel University
- Other Identifier
- 7456; 991014632226904721