Center for Biotechnology and Genomic Medicine Theses and
Dissertationshttp://hdl.handle.net/10675.2/6058712024-03-28T18:29:45Z2024-03-28T18:29:45ZGenomic Predictions in Uterine CancersTran, Lynn Kim Hoanghttp://hdl.handle.net/10675.2/6232692020-05-20T17:54:55Z2020-05-01T00:00:00ZGenomic Predictions in Uterine Cancers
Tran, Lynn Kim Hoang
Introduction: Current uterine cancer classification provides suboptimal treatment stratification and often groups together patients with significant differences in survival outcome and/or response. We used transcriptomic information to devise genomic scores for improved prediction of uterine cancer patient outcomes and validated these scores in our institutional cohorts.
Project 1: In an early iteration of our gene signature discovery pipeline, we developed USC73, a genomic score for uterine serous carcinoma patients, which grouped patients into a low score (lower 66.7 percentile), good prognosis group and a high score (upper 33.3 percentile), poor prognosis group (5-year overall survival: 83.3% and 13.3%, respectively). USC73 predicts survival independently of stage, and can be combined with stage for further resolution of patient survival. Poor survivors have faster-growing tumors and lower rates of complete response to primary therapy.
Project 2: We applied our pipeline to uterine endometrioid carcinoma, the most common histotype of uterine cancer, and developed UEC_IGS, an immune gene score that separates early stage patients into a high lymphocytic infiltration, good prognosis group (IGS 1) and a low lymphocytic infiltration, poor prognosis group (IGS 2). UEC_IGS predicts overall survival independent of grade and treatment. IGS 1 patients have higher levels of CD8+ tumor infiltrating lymphocytes (TILs), more CD45RO+/CD3+ memory T cells, and lower levels of FOXP3+ Tregs compared to IGS 2.
Conclusion: Using transcriptomic data, we can reliably stratify uterine cancer patients into good and poor survival groups. This information can be used to facilitate recruitment of only poor prognosis patients into clinical trials, mitigating some heterogeneity in patient response and allowing clinicians to better identify treatments for patients who will not survive on the current therapy. Additionally, biological functions (e.g. cellular proliferation or immune infiltration) are associated with each genomic score, and these can serve as potential pathways to target for improving the outcome of poor survival groups.
Record is embargoed until 04/28/2022.
2020-05-01T00:00:00ZAPPLICATIONS OF MACHINE LEARNING TO GENOMICS: STUDIES IN TYPE 1 DIABETES AND CANCERTran, Paulhttp://hdl.handle.net/10675.2/6232472020-05-20T17:54:54Z2020-04-01T00:00:00ZAPPLICATIONS OF MACHINE LEARNING TO GENOMICS: STUDIES IN TYPE 1 DIABETES AND CANCER
Tran, Paul
Introduction: A major aim of modern medicine is to translate basic genomics findings using machine learning and other data analysis methods into clinical tests for improving patient care. Herein, I applied machine learning methods to publicly available genetic and genomic data to address three clinical problems in cancer and type 1 diabetes (T1D) research.
Project 1: Cancer classification mostly depends on the anatomic pathology workforce; hence, diagnosis is slow, stepwise, and prone to errors and systemic bias. Using a transcriptome-based cancer classification method, I reconciled the 18% disagreement rate between histology and mutation-based classifier for brain cancer.
Project 2: I applied the same transcriptome-based classification method to lung adenocarcinoma and identified 3 novel subgroups comprising ~30% of lung adenocarcinoma.
Project 3: The estimated genetic heritability of T1D is up to 80%. Identifying those most genetically susceptible to T1D can lead to reduction of the number of islet autoimmunity cases and the number diabetic ketoacidosis episodes. I developed a genetic risk prediction model using neural networks which performs better than currently published methods. I applied model interpretation methods to the neural network and identified important genetic drivers for characterizing T1D molecular subgroups.
Conclusion: These projects are small steps in translating genomic medicine projects to clinical applications but represent a future with more objective and automated tools to aid in clinical decision making.
Record is embargoed until 04/16/2022
2020-04-01T00:00:00ZA novel subnetwork based analysis reveals shared pathways in T-cell mediated autoimmunityPabla, Simarjot Singhhttp://hdl.handle.net/10675.2/6058882020-05-22T17:50:27Z2016-03-01T00:00:00ZA novel subnetwork based analysis reveals shared pathways in T-cell mediated autoimmunity
Pabla, Simarjot Singh
Thymocyte auto-reactivity is an underlying theme of several autoimmune disorders. The precise role of auto-reactive T cells in the initiation and subsequent progression of autoimmune disorders has been studied extensively. However, these disease specific studies ignore pathways that may be in common to several T cell mediated autoimmune pathologies. This can be attributed in part to the shortcomings of traditional gene list based gene expression studies. Here we report a novel method to identify unifying gene expression changes in several autoimmune diseases. In order to uncover pathologically important pathways common to T-cell mediated autoimmune disorders, we used human gene expression data from Multiple Sclerosis, Rheumatoid Arthritis, Juvenile Idiopathic Arthritis and Sjögren’s syndrome. Unlike traditional gene expression analysis, we used jointly active connected subnetwork enrichment to identify subnetworks for each disorder, followed by topological network alignment, which led to identification of shared pathways. We report four pathways shared in these disorders, which include DNA damage response, gonadotropin, innate and adaptive immunity pathways. Importantly, our method did not reveal any common pathways in unrelated diseases. In order to experimentally validate our findings, RNA sequencing of mRNA isolated from salivary glands excised from a murine model of Sjögren’s syndrome was performed. High similarities were observed between Human T-cell mediated autoimmune disorders and Sjögren’s murine model. Collectively, these studies have identified a shared landscape of pathologically significant pathways, including DNA damage response, gonadotropin, innate and adaptive immunity in autoimmune disorders and provide a new methodology to identify common alterations in diseases with similar underlying etiologies.
2016-03-01T00:00:00ZThe Autoimmune Regulator (Aire) Confers Immunosuppressive Properties to Dendritic CellsEisenman, Danielhttp://hdl.handle.net/10675.2/3451362020-08-11T15:04:17Z2007-05-01T00:00:00ZThe Autoimmune Regulator (Aire) Confers Immunosuppressive Properties to Dendritic Cells
Eisenman, Daniel
The Autoimmune regulator (Aire) is a transcription factor that controls expression of self antigens by thymic epithelium and it plays a critical role in the deletion of autoreactive thymocytes and prevention of autoimmunity. Recent studies have reported Aire expression in dendritic cells (DC) located in spleen and lymph nodes, suggesting a role for Aire in extra-thymic tolerance induction. Molecular and functional studies conducted in this dissertation revealed that Aire induction in bone marrow derived DC results in expression of immunosuppressive cytokines and decreased expression of co-stimulatory molecules. Similar results were also obtained from lenti-virus-mediated Aire overexpression in the DC2.4 dendritic cell line. It was further shown that DC from Aire'7' mice exhibited greater antigen presenting function both in vitro and in vivo. These DC were more potent stimulators of T cell proliferation leading to increased IL-2 and IFNy production. These studies suggest that Aire7' DC may play a role in exacerbating the autoimmunity seen in Aire7' mice. DC over-expressing Aire were shown to suppress activation and proliferation of naive T cells and promote activation-induced cell death of activated T cells. Furthermore, we demonstrate that Aire also controls transcription of tissue-specific antigens in DC. These results, together, suggest that Aire plays an important role in the tolerogenic function of DC.
2007-05-01T00:00:00Z