Learning Hybrid Biophysical Neuron Models with Neural ODEs
2026-06-15 • Machine Learning
Machine Learning
AI summaryⓘ
The authors present a new modeling method that combines traditional biophysical neuron models with neural ordinary differential equations (ODEs) to better capture ion channel behaviours that are not well understood or are simplified in existing models. Their approach learns interpretable dynamics directly from voltage data without needing predefined equations. They tested this hybrid model on thousands of ion channels and showed it can generalize to new conditions and reduce computational complexity in neuron simulations. Overall, their method allows researchers to improve neuron models by flexibly adding unknown components while keeping the existing biological mechanisms clear.
biophysical neuron modelsion channelsneural ordinary differential equationsconductance-based modelsgating kineticscurrent-clamp recordingsmulticompartment neuron modelsaxial currentmodel reductionvoltage-dependent dynamics
Authors
Jonas Beck, Michael Deistler, Dóra Viktória Molnár, Jakob H. Macke, Philipp Berens
Abstract
Biophysical neuron models link measurements of neural activity to underlying cellular mechanisms. Yet, a central challenge is that the kinetics of many ion channels are poorly characterized, and practical simplifications -- omitting channels or reducing morphological detail -- introduce systematic gaps between model and biology. Bridging these gaps requires approaches that can flexibly discover unmodeled dynamics while preserving mechanistic interpretability. Here, we introduce a hybrid modeling framework that embeds neural ordinary differential equations into conductance-based biophysical models to capture unknown currents or mis-specified channel kinetics. By parameterizing the neural ODE in terms of voltage-dependent steady-state and time-constant functions, we recover interpretable gating dynamics directly from voltage recordings without assuming a functional form. We show that the hybrid model fits the gating kinetics of 2400 ion channel models and recovers unknown gating dynamics from single current-clamp recordings, generalizing to out-of-distribution stimulus regimes under realistic inputs and parameter misspecification. We also use our method to reduce a multicompartment model of a cortical neuron into a single-compartment hybrid model with a learned axial current, yielding up to an order of magnitude lower computational cost. Together, our results establish a plug-and-play framework for selectively replacing unknown components of conductance-based models with neural ODEs while preserving their mechanistic structure.