Automated inference of disease mechanisms in patient-hiPSC-derived neuronal networks

Human induced pluripotent stem cells (hiPSCs)-derived neuronal networks on multi-electrode arrays (MEAs) are a powerful tool for studying neurological disorders. The electric activity patterns of these networks differ between healthy and patient-derived neurons, reflecting underlying pathology. However, elucidating these underlying molecular mechanisms requires strenuous additional experiments. Computational models can link observable network activity to underlying mechanisms by estimating biophysical model parameters that simulate the experimental observations, but this is challenging. Here, we address this challenge using simulation-based inference (SBI), a machine-learning approach, to automatically estimate all model parameters that can explain network activity. We show how SBI can accurately estimate parameters that replicate the activity of healthy hiPSC-derived neuronal networks, pinpoint molecular mechanisms affected by pharmacological agents, and identify key disease mechanisms in patient-derived neuronal networks. This demonstrates SBI's potential to automate and enhance the discovery of in vitro disease mechanisms from MEA measurements, advancing research with hiPSC-derived neuronal networks.