HIPPIE: A Multimodal Deep Learning Model for Electrophysiological Classification of Neurons

bioRxiv (Cold Spring Harbor Laboratory), Journal Year: 2025, Volume and Issue: unknown

Published: March 15, 2025

Abstract Extracellular electrophysiological recordings present unique computational challenges for neuronal classification due to noise, technical variability, and batch effects across experimental systems. We introduce HIPPIE (High-dimensional Interpretation of Physiological Patterns In recordings), a deep learning framework that combines self-supervised pretraining on unlabeled datasets with supervised fine-tuning classify neurons from extracellular recordings. Using conditional convolutional joint autoencoders, learns robust, technology-adjusted representations waveforms spiking dynamics. This model can be applied clustering diverse biological cultures technologies. validated both in vivo mouse vitro brain slices, where it demonstrated superior performance over other unsupervised methods cell-type discrimination aligned closely anatomically defined classes. Its latent space organizes along gradients, while enabling individual corrected alignment experiments. establishes general systematically decoding diversity native engineered

Language: Английский

Starting a synthetic biological intelligence lab from scratch

Patterns, Journal Year: 2025, Volume and Issue: unknown, P. 101232 - 101232

Published: April 1, 2025

Language: Английский