Link to the University of Pittsburgh Homepage
Link to the University Library System Homepage Link to the Contact Us Form
D-Scholarship @ Pitt Banner and Links to Homepage

Statistical Inference Methods to Identify Tumor Microenvironment Heterogeneity

Zhang, Han (2025) Statistical Inference Methods to Identify Tumor Microenvironment Heterogeneity. Doctoral Dissertation, University of Pittsburgh. (Unpublished)

[img] PDF
Restricted to University of Pittsburgh users only until 31 January 2026.
Download (9MB) | Request a Copy

Abstract

The tumor microenvironment (TME) is a dynamic ecosystem that is continually tested and shaped by the tumor cells. It protects the tumor cells from being attacked by the host immune system and facilitates tumor growth. Therefore, it is of significant interest to understand the cell- cell interplays within the TME to help demystify the mechanisms of immune evasion. However, current computational methods fail to reveal the whole picture of the cell-cell interactions since the activated cell functions in the cell are not fully captured. Several text mining models and factorization analysis methods have been developed to solve this problem; nevertheless, they require strong hidden assumptions that are not met in the context of molecular biology. The goal of this work is to explore a way to quantify and estimate the utility of upregulated functions (specific biological activities that a cell would perform) for cell differentiation and specialization, through which to understand and interpret the communications between cells in a causal relationship. To this end, I developed three aims and address them step by step: 1) scGEM is a nonparametric Bayesian model based on a Dirichlet process to identify correlated gene co- expressing modules within the cell subtypes. The gene modules identified by scGEM are expected to follow cell differentiation path and reflect cellular functions at a higher resolution than the current methods; 2) CRCAtlas uses scGEM to investigate gene modules of colorectal cancer and then employs a pharmacological computational method to locate the ligand receptor pairs that explain such correlations; 3) IOhub is introduced as one of the largest publicly curated databases for immuno-oncology research. By incorporating the gene modules from single cells, IOhub has the potential to discover new biomarkers to predict clinical response to immune checkpoint blockade. Overall, this dissertation contributes to the understanding of cancer immunology and the advancement of precision medicine.

Share

Citation/Export:
Social Networking:
Share |

Details

Item Type: University of Pittsburgh ETD
Status: Unpublished
Creators/Authors:
CreatorsEmailPitt UsernameORCID
Zhang, Hanhansolo.biostat@gmail.comhaz960000-0003-3023-4767
ETD Committee:
TitleMemberEmail AddressPitt UsernameORCID
Thesis AdvisorChen, Lujialuc17@pitt.edu
Committee ChairCooper, Gregorygfc@pitt.edu
Committee MemberLu, Xinghuaxinghua@pitt.edu
Committee MemberOsmanbeyoglu, Hatice Ulkuosmanbeyogluhu@pitt.edu
Committee MemberLu, Binfengbinfeng.lu@hmh-cdi.org
Date: 31 January 2025
Date Type: Publication
Defense Date: 2 December 2024
Approval Date: 31 January 2025
Submission Date: 10 December 2024
Access Restriction: 1 year -- Restrict access to University of Pittsburgh for a period of 1 year.
Number of Pages: 159
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: Tumor Microenvironment, Bioinformatics, Cancer Immunology, Immune Checkpoint Blockade, Biomarker, Cell Cell Communication
Related URLs:
Date Deposited: 31 Jan 2025 14:34
Last Modified: 31 Jan 2025 14:34
URI: http://d-scholarship.pitt.edu/id/eprint/47226

Metrics

Monthly Views for the past 3 years

Plum Analytics

Actions (login required)

View Item View Item