(External) Detection of (emergent) pathogens in North American environmental metagenomes with Machine Learning
Advisor: Susanne Kramer, Environment and Climate Change Canada
Suggested co-advisors: Nicole Ricker
The global rise of antimicrobial resistant (AMR) microorganisms is a current major health concern, largely driven by widespread antimicrobial use (AMU) in both human health and animal production. To combat the spread of AMR, the One Health Framework has been adopted in research contexts to better understand the complex flow of resistance genes and organisms between humans, animals, and the environment. Microbial metagenomes generated from environments across North America present a valuable, yet underutilized resource for investigating questions related to ecosystem function and condition under increasing anthropogenic pressure, such as AMU. Specifically, metagenomes collected across aquatic and terrestrial systems (water, sediment, soil) can be mined to estimate the prevalence of known pathogens, antimicrobial resistance genes (ARGs) and mobile gene elements (MGEs) (e.g., Subirats, 2023). Investigating the diversity, prevalence and distribution of these elements in the North American environment is critical to establish baseline data of the environment’s role as an AMR reservoir and will contribute to One Health AMR risk modelling, which is currently being undertaken at the federal level. We are looking for a student to partner with Environment and Climate Change Canada to mine a comprehensive metagenomic dataset compiled from publicly available databases using machine-learning algorithms to assess patterns of pathogens and ARGs across North American ecozones, land-use features, and population densities. Resultant data insights will help inform priority sites for environmental AMR surveillance strategies and the trained dataset will be used in future projects to enable prediction of emerging threats.
Two-semester project preferred.
Knowledge/Skills
- Necessary: Basic coding skills (R/python), familiarity with CLI, linux
- Ideal: Familiarity with metagenomic dataset mining, machine learning algorithms, cloud computing