Multiscale modelling of progression in Parkinson's disease

Parkinson's disease (PD) is an incurable neurodegenerative condition affecting over ten million people worldwide. It is highly variable in presentation and progression, and whilst disease subtypes have been described at both the clinical and molecular scale, how these relate to one another remain poorly understood. To bridge this gap, we will map the mechanisms of disease progression in PD across multiple biological scales, from cells to circuits, in the same individuals. We will develop a multiscale generative model to characterise how pathophysiological mechanisms interact within and between different levels and allow us to ask how these processes result in the observed clinical syndrome. This proposal would be impossible to achieve working independently, and hinges on a synergistic collaboration between molecular, cellular, clinical and computational neuroscience. It will develop a framework to allow disease types to be defined by their cause, rather than their phenotype, and provide a platform to help dissect mechanistic pathways across scale, and more rapidly develop precision therapies. Delivering tools to integrate data across scale in the same individuals, our vision is to provide an accessible framework, applicable to many other disorders, to allow more rapid translation of targeted treatments that can be started before irreversible loss of tissue has occurred.

Disease progression is a key feature of PD, defined in terms of pathology within a brain region (c.f., symptom severity) and the spread of pathology between regions (c.f., new symptoms). Progression is highly variable, and potential mechanisms span many biological scales: with differences in where the disease begins, pattern of damage across circuits, co-existing pathologies, abnormal neural dynamics, cellular vulnerability and organelle dysfunction all thought to contribute.

This project aims to map the mechanisms of disease progression in Parkinson's disease (PD) across multiple biological scales, from cells to circuits, in the same individuals. Framing PD progression as differences in the cortical burden of disease (cortical dominant = fast; brainstem dominant = slow), we will identify individuals with extreme phenotypes of progression from an existing cohort to:

(i) Define longitudinal changes in brain microstructure in vivo.
(ii) Map the molecular and cellular dynamics using in vitro human models from the same individuals.
(iii) We will integrate data (i) and (ii) to inform a deep generative model, where mechanisms at one scale-quantified via model parameters derived from experimental measures-provide information used by the next. This will characterise how these pathophysiological mechanisms at different scales interact to result in the clinical syndrome observed.

This program will deliver an open framework of integrated resources that can then be used in a translational setting.