Author ORCID Identifier

https://orcid.org/0000-0002-4182-8440

Date of Award

2024

Document Type

Thesis (Ph.D.)

Department or Program

Molecular and Systems Biology

First Advisor

Xiaofeng Wang

Abstract

The three-dimensional organization of the genome is fundamental in regulating gene expression and maintaining cellular function. This organization's complexities, influenced by epigenetic marks and chromatin remodeling complexes, are crucial for understanding genomic regulation. Among these, the SWI/SNF complexes are key, facilitating chromatin accessibility and regulating gene activity across cell types. The first part of my dissertation focuses on SWI/SNF complexes, exploring their role in chromatin remodeling and their impact on 3D genome architecture. Utilizing next-generation sequencing (NGS) techniques, this section investigates the interplay between these complexes and chromatin structure. During my research on the SWI/SNF complex, I was intrigued by the processes governing multi-enhancer interaction, which co-regulates gene expression. Building on this, my dissertation expands the scope by exploring the intrinsic nature of enhancer communities and their broader impact on genome-wide interactions. In the second part, my dissertation shifts to computational modeling, employing deep learning techniques to predict and interpret genomic interactions with unprecedented accuracy. By integrating multi-omic data—including epigenomic features and DNA sequences—this study utilizes deep learning modules such as Convolutional Neural Networks (CNN), Graph Neural Networks (GNNs), and transformers. This approach enables detailed predictions of the dynamic organization of chromatin, surpassing traditional experimental methods. By bridging computational methods with biological inquiries, my thesis aims to contribute new perspectives and tools for investigating the organization of the genome, thereby enriching our understanding and opening up new avenues for research in genomic science.

Download

Share

COinS