Our work combines machine learning, mathematical modelling and experiment to understand human health and disease


Machine learning & AI

Modern experiments and healthcare records make vast amounts of data available. Our work aims to develop new techniques for analysing such complex biomedical datasets. We are particularly interested in linking heterogeneous clinical datasets with molecular profiling and genomic data. We combine statistical analysis with mechanistic modelling to better understand the causes of observed patterns.

Network of collaborations at the University of Southampton

Structure & dynamics of networks

Complex systems can often be represented as networks. Our work explores the structural properties of these networks and the ways in which network architecture relates to system behaviour. We have particular a interest in how social networks affect human health and well-being, the development of network-based representations of data and learning methods that make use of patterns of connectivity.

Neural progenitors derived from mouse ES cells

Stem cell biology

Stem cells are present in most adult tissues, and regulate healthy tissue growth and repair. We use experimental methods and mathematical models to investigate molecular regulation of stem cell fate. Our models include both deterministic and stochastic mechanisms, and we have a particular interest in cell-cell variability and its role in collective decision-making.