Font Size: a A A

Inferring Biological Knowledge of Pathways from an Ontology Fingerprint-derived Gene Network

Posted on:2013-02-09Degree:Ph.DType:Thesis
University:Medical University of South CarolinaCandidate:Qin, TingtingFull Text:PDF
GTID:2458390008486406Subject:Biology
Abstract/Summary:
Biological pathways are the functional units for studying biology at the systems level. Studying biological pathways via systems biology approaches can provide comprehensive insights into the components of pathways, relationships between pathways, and regulatory mechanisms within pathways, thus facilitating hypothesis generation for downstream experimental research as well as the inference of novel therapeutic targets for diseases.;In a previous study, Dr. Zheng's group introduced the concept of the Ontology Fingerprint and proved that using the Ontology Fingerprint can enhance the analysis of high-throughput experimental results. By pairwise comparison of the Ontology Fingerprints of genes, we can quantitatively describe the biological associations between genes which are readily represented by a weighted undirected gene network. In this dissertation, we utilized graphical properties extracted from the Ontology Fingerprint-derived gene Network (OntoFing-Net) to discover novel biological knowledge of pathways, including novel components of pathways, biological relationships between pathways, and signal transductions within pathways.;In Chapter 2, we developed an OntoFing-Net for the yeast genome, and used it to identify novel genes that can potentially modulate biological pathways. We used the yeast sphingolipid pathway as an example and developed a candidate gene prioritization approach based on the principle of guilt-by-association (GBA). The in vivo experiments were performed to validate predicted novel genes for the sphingolipid pathway. In Chapter 3, the biological relationships between pathways were investigated from the human OntoFing-Net. By using shortest-path network analysis, we systematically explored the relationships between the human sphingolipid pathway and various cancer pathways. Finally, in Chapter 4, we incorporated the prior biological knowledge from the OntoFing-Net into Bayesian network analysis. Applying this enhanced Bayesian network analysis, we predicted the structure of a cancer cell type-specific signaling pathway as well as its responses to different biological stimuli using actual experimental data.;These three studies document the applications and advantages of the OntoFing-Net for biological pathway study from different aspects, which facilitate the understanding of pathway modulation mechanism and disease pathogenesis.
Keywords/Search Tags:Biological, Pathways, Gene, Ontology, Network, Ontofing-net
Related items