Computational methods for inferring regulatory mechanisms from sequence and expression variation

With the rapid development and decreasing cost of sequencing technologies, more and more novel genetic variants are being detected. Numerous human genetic and bioinformatic studies associate genotype data to phenotype information and provide increasing number of phenotype-related variants. Despite the large number of associations been detected, we are not even close to a complete understanding of the mechanisms how the genetic variants contribute to phenotypic variation. With the vast majority of the genetic variants from Genome-wide Association Study (GWAS) located on the non-coding region of human genome, it is crucial to understand the gene regulatory mechanisms to be able to interpret the variants. Therefore, the goal of my dissertation is to unveil the molecular mechanisms in the context of human disease using the genetic variants associated with the diseases of interest and also to get a better understanding of gene expression regulation which may in turn improve our interpretation of sequence variants. In this dissertation, Chapter 2 introduces a pipeline through which we associated transcription factors (TFs) with drug response variation. The pipeline involves a novel computational model that predicts TF binding strength for given DNA sequences. Chapter 3 addresses the variant set characterization task where the goal is to rank biological pathways for association with a given set of variants. For this, we developed a computational tool which applies “Random Walk with Restarts” algorithm on a network composed of single-nucleotide polymorphisms (SNP), genes and pathways to associate pathways to a given set of variants. In Chapter 4, we analyze the mechanisms of another group of gene expression regulators: long non-coding RNAs or lncRNAs. With lncRNA-mRNA connections mapped out, non-coding variants can be annotated from more functional categories. To enable this, we modeled the expression of protein coding genes using lncRNAs with potential regulatory mechanisms and created a high confidence lncRNA-mRNA regulatory network.

Name thesis:degree_name

Ph.D.

Level thesis:degree_level

Dissertation

Discipline thesis:degree_discipline

Biophysics & Quant Biology

Grantor

University of Illinois at Urbana-Champaign

Year dc:date

2022

Author dc:creator

• Xie, Xiaoman

Contributors dc:contributor

• Sinha, Saurabh
• Belmont, Andrew S
• Han, Hee-Sun
• Zhao, Sihai Dave

• Gene Regulation
• Variant Interpretation
• Biological Pathways
• Regression Analysis

Statement dc:rights

• Copyright 2021 Xiaoman Xie

Language dc:language

en

Handle dc:identifier

http://hdl.handle.net/2142/113190

OAI identifier oai:identifier

oai:www.ideals.illinois.edu:2142/113190