Our research combines advanced statistical learning and deep learning approaches with high-throughput biological data to reveal new insights into the organisation, evolution, and dynamics of living systems.

By integrating data across molecular, cellular, and population scales, we aim to develop predictive and interpretable models that advance both fundamental understanding and practical applications in bioscience.

Mathematical approaches

Computer mock up of a DNA strand — We are aiming to advance fundamental understanding and practical applications in bioscience.

We develop and apply methods from machine learning, deep learning, Bayesian inference, and network analysis to uncover patterns and structure in large, complex biological datasets. Our work also explores hybrid modelling frameworks that integrate data-driven AI methods with mechanistic models, enabling quantitative prediction and hypothesis generation.

Applications

Our research applies AI and big data methods to genomic, metagenomic, single-cell, and bioimaging data to better understand bacterial populations, microbial communities, and embryonic development. We analyse large-scale genome (including long-read) datasets for bacterial and fungal pathogens and use machine learning to identify genomic features associated with high-risk strains.

These approaches support source attribution for foodborne outbreaks and improve our ability to predict pathogen emergence, transmission routes, and antimicrobial resistance.