Symmetry-informed model inference for active matter


3:30 to 4:30 p.m., Feb. 2, 2024


Speakers:         Jorn Dunkel, Department of Mathematics, MIT

Title:                Symmetry-informed model inference for active matter

Abstract:         Recent experimental advances enable high-resolution observations of biological and synthetic active matter across a wide range of length and time scales. A major challenge is to translate high-dimensional live-imaging data into PDE models that will allow us to predict and understand the emergent collective dynamics seen in experiments. Here, I will describe our current efforts to implement computational frameworks capable of learning interpretable continuum models directly from spatio-temporal tracking data. After outlining theoretical and computational challenges posed by state-of-the-art microscopy and sequencing data, we will show how symmetry concepts and recent algorithmic advances can be combined to construct efficient mode representations and robust inference schemes for automated model discovery. To illustrate the practical potential, we present example applications ranging from the undulatory locomotion of worms and snakes [1] to the collective dynamics of cells [2], active colloids and fish [3].

[1] Cohen, Hastewell, et al., PRL 130: 258402, 2023 [2] Romeo, Hastewell, et al., eLife  10: e68679, 2021 [3] Supekar et al., PNAS 120: e2206994120, 2023