Mathematical & ML Models of Adaptive Stress (Hormesis) in Health and Disease

We are recruiting a PhD student to develop data-driven mathematical models of biphasic stress responses—the “beneficial-at-low-dose, harmful-at-high-dose” patterns seen in ageing and disease (often termed hormesis). You’ll integrate real-world datasets (e.g., exercise training, therapeutic dosing, and clinical/longitudinal cohorts) with mechanistic and statistical modelling to quantify how small stressors (training load, therapy intensity, timing) can optimise outcomes such as VO₂max, cardiometabolic risk, and mortality proxies.

You will:

• Build and compare dose–response and biphasic models (ODE/SDE, state-space, Bayesian hierarchical, causal inference, interpretable ML).
• Develop tools to discover turning points (benefit→risk transitions) and personalise prescriptions.
• Validate models against open datasets and prospective data; produce reproducible code and visualisations.

You bring:

• Background in applied maths, statistics, computer science, or related field.
• Solid skills in programming; familiarity with ODE or probabilistic modelling, optimisation, or ML.
• Interest in translating methods across domains (control, pharmacometrics, reliability) into biomedicine.

Why this PhD:

• Work at the interface of mechanistic modelling and machine learning with immediate impact on exercise and therapy design.
• Publish in both methodological and application venues; collaborate with clinicians, sport scientists and other mathematicians/data scientists.