AI+Biology
Developing artificial intelligence (AI) models that extract insight from complex biological data to support discovery, prediction, and decision-making across health, agriculture, and the environment.
Biology is full of complexity, from how infections spread to how microbes shape health and ecosystems. The AI + biology theme applies advanced AI methods to make sense of this complexity, combining machine learning, reasoning over biological networks, and language-based models with domain expertise. Our goal is to turn minimal or scattered data into actionable insight—whether predicting the impact of new pathogens, accelerating vaccine discovery, or guiding patient-specific interventions—ensuring that AI contributes to timely and transparent decisions in medicine, veterinary science, agriculture, and environmental management.
Our research
We design technically rigorous, interpretable, and data-efficient AI methods to address fundamental biological challenges. Current strands include:
- Pathogen prediction – inferring host range, transmission routes, and pathogenicity of novel pathogens from minimal sequence and contextual data
- Biological interactions – applying graph-based learning to uncover beneficial microbial, host, and environmental relationships that improve health and food security
- Personalised interventions – integrating structured and unstructured biological data to guide patient-specific care and research
- Safe and transparent AI – applying verification, uncertainty estimation, and robustness analysis to ensure trustworthy biomedical tools
- Generative and causal modelling – accelerating vaccine and drug discovery in data-scarce settings
- Language models for biology – tuning large language models to synthesise knowledge from biological literature and databases
- Active learning for experiments – developing explainable AI to prioritise and design biological experiments under uncertainty.
Our work has been cited in national advisory documents and UK Research and Innovation strategic planning, and has produced case studies such as a big data approach to understanding disease transmission.
People
The AI + biology theme is led by researchers with expertise in AI, computational biology, and systems-level biology:
- Dr Antony McCabe
- Dr Yi Dong
- Dr Maya Wardeh.
Partnerships and collaborations
Our research is highly interdisciplinary, working closely with colleagues across the Institute of Infection, Veterinary and Ecological Sciences and the Computational Biology Facility. We also contribute to global resources such as the Eukaryotic Pathogen Vector and Host Database (VEuPathDB) project, supporting communities in infection biology and public health.
Outputs and impact
AI + biology accelerates the path from discovery to intervention. Our models can turn minimal data on a new pathogen into early warnings, support vaccine development, and provide resources for clinical and population-level decisions such as transplantation. The impact extends across healthcare, veterinary science, agriculture, and environmental management, where transparent and interpretable AI is essential for responsible decision-making. By combining data-driven learning with domain expertise, our outputs influence both scientific practice and national strategy in health and biosecurity.