2026-05-22T05:12:01Zhttps://keep.lib.asu.edu/oai/request

oai:keep.lib.asu.edu:node-2023552025-08-18T22:22:09Zoai_pmh:alloai_pmh:repo_items

202355 https://hdl.handle.net/2286/R.2.N.202355 http://rightsstatements.org/vocab/InC/1.0/ All Rights Reserved 2025 147 pages Doctoral Dissertation Academic theses en Robles Fernandez, Angel Luis Upham, Nathan Sterner, Beckett Franz, Nico Suzuki, Taichi Arizona State University Partial requirement for: Ph.D., Arizona State University, 2025 Field of study: Biology To accelerate the prediction of pathogen spillover, this work presents a framework for predicting host-pathogen interactions through ecological and evolutionary covariates. My research focuses on machine learning (ML) workflows for the high-throughput data structuring and modeling of host-parasite interactions from published literature and open databases. It has been developed in three branches. First, I start with the automation of host-parasite data extraction. Second, I establish a theoretical framework relating genetic diversity to environmental suitability. This theoretical framework, along with a set of well-defined assumptions, allow me to hypothesize about the spatial distribution of parasite species richness and to predict host-parasite interactions through host ecological covariates. Finally, the ML framework for predicting host-parasite interactions from these covariates is presented. Each part of this approach solves a specific task. However, the final products are assessment tools associated with host-parasite interactions, predicted pathogen interaction networks, and model interpretations (i.e., variable importance metrics and partial dependent plots). This integrative framework will provide open host-parasite knowledge to increase societical knowledge about zoonotic disease processes. Ecology Machine learning Preditive biogeography Ecointeraction: A Machine Learning Framework to Predict Host-parasite Interactions Through Ecological and Evolutionary Covariates