HIPPIE: a generative model for electrophysiological analysis across species, technologies, and modalities
10 affiliations
- 1Department of Biomolecular Engineering, University of California Santa Cruz, Santa Cruz, CA, USA
- 2Genomics Institute, University of California Santa Cruz, Santa Cruz, CA, USA
- 3Live Cell Biotechnology Discovery Lab, University of California Santa Cruz, Santa Cruz, CA, USA
- 4Department of Molecular, Cellular and Developmental Biology, University of California Santa Cruz, Santa Cruz, CA, USA
- 5Department of Electrical and Computer Engineering, University of California Santa Cruz, Santa Cruz, CA, USA
- 6Biotechnology Program, Berkeley City College, Berkeley, CA, USA
- 7Department of Statistics & Data Science, Yale University, New Haven, CT, USA
- 8Wu Tsai Institute, Yale University, New Haven, CT, USA
- 9Genomics Institute, University of California Santa Cruz, Santa Cruz, CA, USA. [email protected]
- 10Live Cell Biotechnology Discovery Lab, University of California Santa Cruz, Santa Cruz, CA, USA. [email protected]
Abstract
Neuronal classification from extracellular electrophysiological recordings is challenging due to intrinsic waveform variability, noise, and technical differences across experiments, technologies, and species. We introduce HIPPIE (High-dimensional Interpretation of Physiological Patterns In Intercellular Electrophysiology), a deep learning framework that combines self-supervised pretraining on unlabeled datasets with supervised fine-tuning to classify neurons from extracellular recordings. Using conditional convolutional joint autoencoders, HIPPIE learns technology-adjusted representations of waveforms and spiking dynamics. Here we show, across mouse, rat, and macaque recordings, that HIPPIE classifies cell types competitively with existing methods while additionally supporting generative analyses that discriminative models cannot perform, including counterfactual decoding of electrophysiological signals under changed experimental conditioning, cross-species latent interpolation, and a cross-modal analysis revealing that spike-timing modalities and waveform morphology encode largely independent dimensions of neuronal identity. HIPPIE is available as both a Python package and a coding-free web application, providing a unified framework for multimodal neuronal classification across technologies, experimental conditions, and species.
Related
- Uncovering cognitive-motor adaptation to body visualization in immersive technologies trough their behavioral and electrophysiological correlates
- Speciesformer learns conserved cellular states for cross-species generative virtual cell modeling
- Blinded by motion: Electrophysiological decoupling of time dilation and confidence
- Electrophysiological Correlates of Maladaptive Daydreaming: An Exploratory Resting-State EEG Study
- Menstrual cycle and hormonal contraceptive intake alter the electrophysiological signature of selective auditory spatial attention
BibTeX
@article{gonzalezferrer2026hippie,
title = {{HIPPIE: a generative model for electrophysiological analysis across species, technologies, and modalities}},
author = {Jesus Gonzalez-Ferrer and Julian Lehrer and Bruno Alvarez-Esteban and Avelina Moreno-Ochando and H. Schweiger and Jinghui Geng and Luiz F. S. Eugênio dos Santos and Sebastian Hernandez and Francisco Reyes and Jessica L. Sevetson and Aidan Schneider and Sofie R. Salama and Mircea Teodorescu and David H. Haussler and Mohammed A. Mostajo-Radji},
journal = {Nature Communications},
year = {2026},
month = aug,
doi = {10.1038/s41467-026-76939-w},
url = {https://doi.org/10.1038/s41467-026-76939-w},
note = {CC BY},
}
RIS
TY - JOUR
TI - HIPPIE: a generative model for electrophysiological analysis across species, technologies, and modalities
AU - Jesus Gonzalez-Ferrer
AU - Julian Lehrer
AU - Bruno Alvarez-Esteban
AU - Avelina Moreno-Ochando
AU - H. Schweiger
AU - Jinghui Geng
AU - Luiz F. S. Eugênio dos Santos
AU - Sebastian Hernandez
AU - Francisco Reyes
AU - Jessica L. Sevetson
AU - Aidan Schneider
AU - Sofie R. Salama
AU - Mircea Teodorescu
AU - David H. Haussler
AU - Mohammed A. Mostajo-Radji
PY - 2026
DA - 2026/08/21
JO - Nature Communications
DO - 10.1038/s41467-026-76939-w
UR - https://doi.org/10.1038/s41467-026-76939-w
AB - Neuronal classification from extracellular electrophysiological recordings is challenging due to intrinsic waveform variability, noise, and technical differences across experiments, technologies, and species. We introduce HIPPIE (High-dimensional Interpretation of Physiological Patterns In Intercellular Electrophysiology), a deep learning framework that combines self-supervised pretraining on unlabeled datasets with supervised fine-tuning to classify neurons from extracellular recordings. Using conditional convolutional joint autoencoders, HIPPIE learns technology-adjusted representations of waveforms and spiking dynamics. Here we show, across mouse, rat, and macaque recordings, that HIPPIE classifies cell types competitively with existing methods while additionally supporting generative analyses that discriminative models cannot perform, including counterfactual decoding of electrophysiological signals under changed experimental conditioning, cross-species latent interpolation, and a cross-modal analysis revealing that spike-timing modalities and waveform morphology encode largely independent dimensions of neuronal identity. HIPPIE is available as both a Python package and a coding-free web application, providing a unified framework for multimodal neuronal classification across technologies, experimental conditions, and species.
SN - 2041-1723
ER -