This interdisciplinary PhD explores how artificial intelligence (AI) can make viral genomic data accessible, interpretable, and actionable, especially in low- and middle-income countries (LMICs) where technical resources are limited. The project will use small language models (SLMs), lightweight AI tools, to generate plain-language summaries from viral genome databases, including rabies and influenza, removing the need for specialist knowledge or technical jargon. Leveraging the soon-to-be-released Virus Genome Toolkit (V-gTK), the SLMs will enable real-time exploration of mutation hotspots, functional impacts, and epidemiological patterns. Outputs will be tailored for different stakeholders - laboratory staff, policymakers, and public or veterinary health practitioners- helping bridge sectoral silos and support One Health collaboration. Embedded social science evaluation will ensure the AI tools are usable, interpretable, and equitable, with stakeholder feedback actively shaping design and deployment. The project will deliver a prototype SLM system and practical guidance for its adoption, enabling LMIC partners to harness AI-enhanced pathogen genomics. By combining cutting-edge AI with genomic surveillance and stakeholder engagement, this work aims to improve outbreak preparedness and inform data-driven decision-making globally.
Seeking candidates with a strong computational background who is passionate and intrinsically motivated by the social, ethical, and equity aspects of technology.