Date of Award:

5-2026

Document Type:

Dissertation

Degree Name:

Doctor of Philosophy (PhD)

Department:

Plants, Soils, and Climate

Committee Chair(s)

Matt Yost

Committee

Matt Yost

Committee

Matthew Robbins

Committee

Yoshimitsu Chikamoto

Committee

Curtis Dyreson

Committee

Soukaina Filali Boubrahimi

Abstract

Diseases caused by harmful microorganisms present serious challenges to agriculture, food security, and natural ecosystems, leading to significant economic losses worldwide. Understanding how these microorganisms interact with their hosts at molecular level is essential for developing effective and sustainable disease management strategies. The research summarized here focuses on using computational approaches to investigate host-pathogen interactions, with particular emphasis on protein-protein interactions that play a key role in disease development. 

With the rapid growth of genomic and proteomic data, extracting meaningful biological insights has become increasingly complex. This work addresses that challenge by developing multiple computational models to predict protein-protein interactions and applying them to a representative plant disease system (Alfalfa-Bacterial stem blight). Functional analyses, including gene ontology enrichment, pathway analysis, and subcellular localization prediction, were integrated to better understand the biological roles of proteins involved in these interactions and to identify key molecular components associated with disease progression. 

To ensure accessibility and long-term usability of the generated data, a dedicated database was developed to store predicted protein interactions along with their functional annotations. This resource enables researchers to explore interaction networks and molecular features of disease-causing bacteria in a user-friendly manner. In addition, a standalone, homology-based software package was developed to predict protein-protein interactions across multiple biological systems, including plant–pathogen, human–virus, human–bacteria, and animal–pathogen interactions, thereby reducing the time and expertise required for such analyses. 

Since the cellular location of proteins strongly influences their function, this work also developed species-specific machine learning models for predicting the subcellular localization of proteins in legume crops. A deep learning–based web server was implemented to provide accurate and accessible predictions. Furthermore, computational structural approaches were applied to identify natural compounds as potential inhibitors of fungal diseases in soybean. Large-scale virtual screening, molecular docking, and simulation studies revealed both common and disease-specific compounds, highlighting environmentally friendly alternatives to chemical fungicides. 

Overall, this research provides computational tools, databases, and predictive models that support future experimental studies and contribute to more precise and sustainable approaches to disease management in agriculture.

Download

Available for download on Thursday, May 01, 2031

Share

COinS

Copyright for this work is retained by the student. If you have any questions regarding the inclusion of this work in the Digital Commons, please email us at .