Integration of bioinformatics approaches to study transcriptome patterns associated with complex traits at the gene and isoform levels

Zhang, Pan

Permalink

https://hdl.handle.net/2142/110638 Copy

Description

Title

Integration of bioinformatics approaches to study transcriptome patterns associated with complex traits at the gene and isoform levels

Author(s)

Zhang, Pan

Issue Date

2021-03-23

Director of Research (if dissertation) or Advisor (if thesis)

Rodriguez-Zas, Sandra Luisa

Doctoral Committee Chair(s)

Rodriguez-Zas, Sandra Luisa

Committee Member(s)

• Caetano-Anolles, Gustavo
• Villamil, Maria Bonita
• Yan, Huihuang

Department of Study

Illinois Informatics Institute

Discipline

Informatics

Degree Granting Institution

University of Illinois at Urbana-Champaign

Degree Name

Ph.D.

Degree Level

Dissertation

Date of Ingest

2021-09-17T02:34:19Z

Keyword(s)

RNA-seq, differential expression, alternative splicing, co-expression network, Gaussian Markov random fields

Abstract

The study of the molecules associated with complex traits such as diseases can advance the understanding of traits and help identify biomarkers and potential remedies. In this study, high-throughput RNA-seq technology, statistical models and bioinformatics approaches were used to profile the transcriptome and uncover the underlying molecular mechanisms associated with activity addiction and opioid-induced hyperalgesia with brain region or central nervous system region dependencies considered. Functional category enrichment, reconstruction of molecular interactions, and regulatory networks added systematic insights into the dysregulation of the transcriptome. In addition, co-expression networks at various aggregate levels (e.g., gene family, gene, or transcript isoform) were estimated using Gaussian Markov random fields based on the abundance of mRNA. Multiple criteria were used to benchmark the inferred networks against the pathways in Kyoto Encyclopedia of Genes and Genomes reference. Our findings suggest that the inference of networks using granular information can enhance the network performance, especially when high splicing variation is involved.

Graduation Semester

2021-05

Type of Resource

Thesis

Permalink

http://hdl.handle.net/2142/110638

Copyright and License Information

Copyright 2021 Pan Zhang

Owning Collections