Date of Completion

7-20-2018

Embargo Period

7-19-2021

Keywords

Functional Pathway Analysis, Target genes, Enrichment analysis Differentiation Expression, Data Integration, Cancer Signaling Pathways

Major Advisor

Dong-Guk Shin

Associate Advisor

Charles Giardina

Associate Advisor

Jinbo Bi

Associate Advisor

Chun-Hsi Huang

Field of Study

Computer Science and Engineering

Degree

Doctor of Philosophy

Open Access

Open Access

Abstract

Recent studies have suggested a key role of activities related to target genes of transcription factors (TFs) in cancer signaling pathways. Molecular pathway curation includes mostly upstream signals which are well-established in the literature, but the activities that are downstream of TFs are not well understood. The purpose of this thesis is to design a computational method for exploring TF target genes in a given molecular pathway. We then use this data to stratify patients and relate TF activity to patient survival.

Transcriptome data can provide a detailed readout of events taking place within a cancer, but new computational tools are needed to extract maximal biological insights from this data. The “BioTarget” pipeline incorporates ChIP-seq data into cellular pathway analysis to infer TF activity from RNA-seq data. More specifically, it relates the expression of TF target genes (based on ChIP-seq data) with the status of upstream signaling components to quantify context-specific TF activity. This mining tool also allowed us to stratify TF binding with activated or suppressed states of these targets. Two case studies were analyzed utilizing the BioTarget pipeline.

First, we evaluated the Th1 and Th2 responses in cancers based on TBX21 and GATA3 pathway activity, respectively. Low Th1/Th2 activity ratios were associated with poorer survival of stomach and breast cancer patients, whereas an unbalanced Th1/Th2 response correlated with a poorer survival of colon cancer patients. Applying our computational tool to the BCL6 TF, a TF with a lesser-known influence on solid cancers, we found patients with higher BCL6 activity had significantly improved survival.

Second, we explored target genes of TFs in three molecular pathways of cell proliferation (TCF7, TCF7L2 and LEF1 in WNT; CREB1, MYC, JUN and FOS in MAPK; STAT3 in IGF). These TFs have well-established roles in cell proliferation and immunosuppression. We used BioTarget to explore the oncogenic functions of these TFs by identifying their target genes in specific cancer types. We discovered that the functions of these TFs were dominantly activated in certain cancer cohorts. The results of two case studies provided new insight into pathway signals of cell proliferation and cell differentiation in cancer.

COinS