Date of Award

Spring 5-10-2025

Level of Access Assigned by Author

Open-Access Thesis

Degree Name

Doctor of Philosophy (PhD)

Department

Biomedical Sciences

First Committee Advisor

Clarissa Henry

Second Committee Member

Joshua Kelley

Third Committee Member

Gregory Cox

Additional Committee Members

Terry Yoo

Romain Madelaine

Abstract

Biology is a vast field filled with complex systems and mechanisms. The advancement of technologies, such as microscopes and gene-editing tools, has further revolutionized the field and generated more complex and untapped datasets. Although current technologies and tools have improved data acquisition and experimental throughput, we still lack a comprehensive understanding of many critical cellular functions. Moreover, conventional methods that rely on manual data analysis have been linked with biases and inconsistencies, reported by the scientific communities. Consequently, there is growing recognition of the need for unbiased quantitative approaches to analyzing data. While investigating muscular dystrophy lines in zebrafish using birefringence imaging, we faced several challenges: human errors and time-consuming tasks. To address the challenges, we developed image analysis pipelines in MATLAB that accelerated the workflow more than 20-fold than conventional methods. These pipelines were able to annotate and quantify zebrafish muscle despite the presence of background noises. We also developed additional features on existing software, HeartBeat, to improve accuracy and accelerate the process of calculating heartbeats and arrhythmias in dystrophic zebrafish. The new software, DanioHeartBeats, is stand-alone software that allows users to upload raw files, adjust video brightness, and correct mislabeling with ease. In another project that focuses on cell polarity in yeast, we built ShmooOme, a high-resolution spatial mapping library for polarized proteins, using novel computational tools to help mapping the precise sequence of events that lead to cell polarity. ShmooOme helps characterize the complex interactions of multiple systems behind the cell polarity by analyzing the polarity protein distributions within the shmoo tips using principal component analysis and hierarchical clustering to identify cellular systems. In our final work in prostate cancer research, we developed ordinary differential equation (ODE) models of AR and PARP7 to help improve our understanding of experimental results, eight-hour degradation delay of transcriptional genes in response to RBN2397. Our mathematical models suggested that the concentration of AR plays a role in the degradation delay observed in the transcripts in the RBN treatment. Additionally, the models suggested that PARP7 ADP ribosylation of AR occurs when AR is bound to DNA, which also affects AR-mediated gene expression level.

Download

Share