Amsterdam Neuroscience

Presentation by Kyle Wedgwood - University of Exeter, United Kingdom 

Kyle Wedgwood is an Associate Professor in the Department of Mathematics and Statistics, housed with the Living Systems Institute of the University of Exeter (United Kingdom). He works primarily in the Faculty of Environment, Science and Economy, but works closely with the Faculty of Health and Life Science. In his research, he applies techniques from mathematical modelling (dynamical systems theory, bifurcation analysis) to understand how networks of cells come together to form biological networks that can perform functional tasks. He is particularly interested in spatio-temporal patterns of neural activity in the brain and their role in memory and spatial navigation, and the synchronisation of electrical activity amongst the insulin-secreting beta cells in the pancreas.  

Presentation: A Closed-loop Framework for Model Discrimination and Parametrisation Using Optimal Control

The ability to accurately predict the response of an observed system to a known input is a fruitful first step to accurately control the system’s dynamics. Despite the recent advances in fully data-driven algorithms, the most interpretable way to reach this goal is through mechanistic mathematical modeling. It is not always possible or straightforward to develop a description of a system of interest from first principles. Often one has to rely on empirical, effective models, with the drawback of having many potential candidates competing at explaining some phenomenon. In such a case, there is often no simple way to choose which one provides the best description as these models might often give similar predictions. In this work, we leverage optimal control and propose a closed-loop iterative method to choose among a set of candidate models the one that most accurately predict an observed system. We assume that one has the ability to apply control to an input of the observed system and access to measurements of its response. Our approach is then to identify the input control that maximally discriminates the response of the candidate models, allowing us to determine which model is best by comparing such responses with the observed data. We first demonstrate our proposed iterative scheme in numerical simulations before applying it during an electrophysiology experiment, successfully discriminating between three different models for photocurrents produced via opsin dynamics. We anticipate that, given its general nature, our scheme is well suited for many other biological and non-biological applications.

Program

Participation is free of charge.

Registration here