Mathematics & statisticsMathematics of Life Sciences

Marine Science

The Marine Modelling Group focusses on applications of mathematics and statistics to the modelling of life in the ocean, and how ocean health is affected by climate change and human exploitation, especially fisheries. Our research involves the translation of empirically-based narratives on the functioning of marine ecosystems into new mathematical and statistical model formulations. These are generally based either on systems of ordinary or partial differential equations, or approaches that merge process-based and machine-learning methods in innovative combinations. Current research includes computational approaches to uncertainty propagation in complex models, process-informed deep learning, and development of effective decision-support systems. Our approaches have contributed to the advancement of fundamental mathematical/statistical theories in ecology.

Group members:

Variation & Selection

A common practice in population dynamic modelling is to assign individuals to compartments and represent their traits by the respective compartmental means. Depending on the trait in question, this use of means can be anything from very reasonable to very problematic. Notoriously difficult problems arise when a trait is subject to selection in such a way that its mean changes over time, rendering homogeneous models inappropriate. In the Variation and Selection group, we develop and analyse deterministic and stochastic mathematical models, which we use in the design and analysis of experimental and observational studies to infer population heterogeneity among other parameters. We use Bayesian statistics, computational statistics, and specialised latent variable models for high-dimensional data. Studies currently being conducted include the spread of epidemics, the growth of bacterial populations, methylation patterns of DNA, and population genetics.

A shared interest group with the same name (VaS-SIG) has been created to provide a coherent, collaborative, environment that channels what individual researchers normally do towards a more ambitious, shared, research aim and the development of joint large-scale funding applications. VaS-SIG can be joined by completing this simple form.

Group members:

Epidemiological Modelling

The Epidemiological Modelling Group uses statistical, mathematical and computing tools to study the emergence, spread, impact and ways to control epidemics across the OneHealth spectrum.  OneHealth is an integrated, unifying approach that aims to sustainably balance and optimise the health of people, animals, plants and ecosystems. Our work ranges from surveillance through prediction of pest and disease spread, analysing the impact of vaccination and other control measures, to incorporating human behaviour and economics, using classical and Bayesian parameter estimation, deterministic and stochastic models, networks, and AI and machine learning. We work on such diseases as malaria, dengue, HPV, influenza and COVID-19, and on pests such as Emerald Ash Borer and bark beetles, and we widely collaborate within the UK and with countries like Brazil, Malaysia, Thailand, and India.

Group members: