Machine-learning-based identification of factors that influence molecular virus-host interactions

Chai, Haiting (2022) Machine-learning-based identification of factors that influence molecular virus-host interactions. PhD thesis, University of Glasgow.

Full text available as:
[thumbnail of 2022chaiphd.pdf] PDF
Download (14MB)

Abstract

Viruses are the cause of many infectious diseases such as the pandemic viruses: acquired immune deficiency syndrome (AIDS) and coronavirus disease 2019 (COVID-19). During the infection cycle, viruses invade host cells and trigger a series of virus-host interactions with different directionality. Some of these interactions disrupt host immune responses or promote the expression of viral proteins and exploitation of the host system thus are considered ‘pro-viral’. Some interactions display ‘pro-host’ traits, principally the immune response, to control or inhibit viral replication. Concomitant pro-viral and pro-host molecular interactions on the same host molecule suggests more complex virus-host conflicts and genetic signatures that are crucial to host immunity. In this work, machinelearning-based prediction of virus-host interaction directionality was examined by using data from Human immunodeficiency virus type 1 (HIV-1) infection. Host immune responses to viral infections are mediated by interferons(IFNs) in the initial stage of the immune response to infection. IFNs induce the expression of many IFN-stimulated genes (ISGs), which make the host cell refractory to further infection. We propose that there are many features associated with the up-regulation of human genes in the context of IFN-α stimulation. They make ISGs predictable using machine-learning models. In order to overcome the interference of host immune responses for successful replication, viruses adopt multiple strategies to avoid being detected by cellular sensors in order to hijack the machinery of host transcription or translation. Here, the strategy of mimicry of host-like short linear motifs (SLiMs) by the virus was investigated by using the example of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2). The integration of in silico experiments and analyses in this thesis demonstrates an interactive and intimate relationship between viruses and their hosts. Findings here contribute to the identification of host dependency and antiviral factors. They are of great importance not only to the ongoing COVID-19 pandemic but also to the understanding of future disease outbreaks.

Item Type: Thesis (PhD)
Qualification Level: Doctoral
Additional Information: This work was funded by China Scholarship Council under Grant 201706620069.
Colleges/Schools: College of Medical Veterinary and Life Sciences > School of Infection & Immunity
Supervisor's Name: Robertson, Prof. David L., Hughes, Dr. Joseph and Gu, Dr. Quan
Date of Award: 2022
Depositing User: Theses Team
Unique ID: glathesis:2022-82931
Copyright: Copyright of this thesis is held by the author.
Date Deposited: 08 Jun 2022 12:24
Last Modified: 08 Jun 2022 12:33
Thesis DOI: 10.5525/gla.thesis.82931
URI: https://theses.gla.ac.uk/id/eprint/82931
Related URLs:

Actions (login required)

View Item View Item

Downloads

Downloads per month over past year