ElectroPhysiomeGAN: Generation of Biophysical Neuron Model Parameters from Recorded Electrophysiological Responses
Curation statements for this article:-
Curated by eLife
eLife assessment
The study by Kim et al. is a valuable contribution to the topic of obtaining good channel conductance parameters from electrophysiological recordings. While promising in its ability to rapidly construct newly fitted models using generative adversarial networks, the approach is incompletely described and the generated models often substantially deviate from the dynamics observed empirically. The comparison with existing multi-objective optimization methods is also incomplete.
This article has been Reviewed by the following groups
Listed in
- Evaluated articles (eLife)
Abstract
Recent advances in connectomics, biophysics, and neuronal electrophysiology warrant modeling of neurons with further details in both network interaction and cellular dynamics. Such models may be referred to as ElectroPhysiome, as they incorporate the connectome and individual neuron electrophysiology to simulate neuronal activities. The nervous system of C. elegans is considered a viable framework for such ElectroPhysiome studies due to advances in connectomics of its somatic nervous system and electrophysiological recordings of neuron responses. In order to achieve a simulated ElectroPhysiome, the set of parameters involved in modeling individual neurons need to be estimated from electrophysiological recordings. Here, we address this challenge by developing a novel deep generative method called ElectroPhysiomeGAN (EP-GAN), which once trained, can instantly generate parameters associated with the Hodgkin-Huxley neuron model (HH-model) for neurons with graded potential response. The method combines Generative Adversarial Network (GAN) architecture with Recurrent Neural Network (RNN) Encoder and can generate an extensive number of parameters (>170) given the neuron’s membrane potential responses and steady-state current profiles. We validate our method by estimating HH-model parameters for 200 synthetic neurons with graded membrane potential followed by 9 experimentally recorded neurons (where 6 of them newly recorded) in the nervous system of C. elegans . Compared to other methods, EP-GAN is advantageous in both accuracy of generated parameters and inference speed. In addition, EP-GAN preserves performance when provided with incomplete membrane potential responses up to 25% and steady-state current profiles up to 75%. EP-GAN is designed to leverage the generative capability of GAN to align with the dynamical structure of HH-model, and thus able to achieve such performance.
Article activity feed
-
-
-
eLife assessment
The study by Kim et al. is a valuable contribution to the topic of obtaining good channel conductance parameters from electrophysiological recordings. While promising in its ability to rapidly construct newly fitted models using generative adversarial networks, the approach is incompletely described and the generated models often substantially deviate from the dynamics observed empirically. The comparison with existing multi-objective optimization methods is also incomplete.
-
Reviewer #1 (Public Review):
The manuscript describes a GAN-based approach that generates parameters for HH-like channels for multiple C. Elengans neurons. The network is trained on generated data to produce parameter sets that, on the one hand, reproduce voltage responses and IV curves, and on the other hand, are indistinguishable from the ground truth parameters, as tested by the discriminator. It is then shown that these generated parameter sets lead to reasonable reproductions of the recorded responses (but see the section "weaknesses" below for some reservations).
Strengths:
In itself, I find the methodology of high interest, particularly in that it can generate parameter sets to construct models of new recordings at a very low computational cost.
Weaknesses:
Nevertheless, I believe there are some weaknesses in the evaluation of …
Reviewer #1 (Public Review):
The manuscript describes a GAN-based approach that generates parameters for HH-like channels for multiple C. Elengans neurons. The network is trained on generated data to produce parameter sets that, on the one hand, reproduce voltage responses and IV curves, and on the other hand, are indistinguishable from the ground truth parameters, as tested by the discriminator. It is then shown that these generated parameter sets lead to reasonable reproductions of the recorded responses (but see the section "weaknesses" below for some reservations).
Strengths:
In itself, I find the methodology of high interest, particularly in that it can generate parameter sets to construct models of new recordings at a very low computational cost.
Weaknesses:
Nevertheless, I believe there are some weaknesses in the evaluation of the models that should be addressed before the quality of the methodology can be fully assessed. Firstly, at the methodological level, the authors should provide more clarity on the inverse gradient operation they use, as opposed to just simulating the models, as such an inversion depends not only on the parameters but also on the state of the model. How the state is obtained remains unclear here. Secondly, in the evaluation of their models, the authors could provided more information such as IV curves, as whether these would be accurate is difficult to visually infer from their figures. Thirdly, the authors do not address the question of whether all obtained parameter sets are stable when simulated over longer times, while their figures do include hints that this might not be the case for at least some of their models (e.g. voltage traces that do not converge back to the equilibrium after the stimulus, but rather seem to diverge).
-
Reviewer #2 (Public Review):
Summary:
Generating biophysically detailed computational models that capture the characteristic physiological properties of biological neurons for diverse cell types is an important and difficult problem in computational neuroscience. One major challenge lies in determining the large number of parameters of such models, which are notoriously difficult to fit into experimental data. Thereby, the computational and energy costs can be significant. The study 'ElectroPhysiomeGAN: Generation of Biophysical Neuron Model Parameters from Recorded Electrophysiological Responses' by Kim et al. describes a computationally efficient approach for predicting model parameters of Hodgkin-Huxley neuron models using Generative Adversarial Networks (GANs) trained on simulation data. The method is applied to generate models for …
Reviewer #2 (Public Review):
Summary:
Generating biophysically detailed computational models that capture the characteristic physiological properties of biological neurons for diverse cell types is an important and difficult problem in computational neuroscience. One major challenge lies in determining the large number of parameters of such models, which are notoriously difficult to fit into experimental data. Thereby, the computational and energy costs can be significant. The study 'ElectroPhysiomeGAN: Generation of Biophysical Neuron Model Parameters from Recorded Electrophysiological Responses' by Kim et al. describes a computationally efficient approach for predicting model parameters of Hodgkin-Huxley neuron models using Generative Adversarial Networks (GANs) trained on simulation data. The method is applied to generate models for 9 non-spiking neurons in C. elegans based on electrophysiological recordings. While the generated models capture the responses of these neurons to some degree, they generally show significant deviations from the empirically observed responses in important features. While interesting, in its current form, the method has not been demonstrated to generate models that faithfully capture empirically observed responses.
Strengths:
The authors work on an important and difficult problem. A noteworthy strength of their approach is that once trained, the GANs can generate models from new empirical data with very little computational effort. The generated models reproduce the average voltage during current injections reasonably well.
Weaknesses:
Major 1: While the models generated with EP-GAN reproduce the average voltage during current injections reasonably well, the dynamics of the response are not well captured. For example, for the neuron labeled RIM (Figure 2), the most depolarized voltage traces show an initial 'overshoot' of depolarization, i.e. they depolarize strongly within the first few hundred milliseconds but then fall back to a less depolarized membrane potential. In contrast, the empirical recording shows no such overshoot. Similarly, for the neuron labeled AFD, all empirically recorded traces slowly ramp up over time. In contrast, the simulated traces are mostly flat. Furthermore, all empirical traces return to the pre-stimulus membrane potential, but many of the simulated voltage traces remain significantly depolarized, far outside of the ranges of empirically observed membrane potentials. While these deviations may appear small in the Root mean Square Error (RMSE), the only metric used in the study to assess the quality of the models, they likely indicate a large mismatch between the model and the electrophysiological properties of the biological neuron.
Major 2: Other metrics than the RMSE should be incorporated to validate simulated responses against electrophysiological data. A common approach is to extract multiple biologically meaningful features from the voltage traces before, during and after the stimulus, and compare the simulated responses to the experimentally observed distribution of these features. Typically, a model is only accepted if all features fall within the empirically observed ranges (see e.g. https://doi.org/10.1371/journal.pcbi.1002107). However, based on the deviations in resting membrane potential and the return to the resting membrane potential alone, most if not all the models shown in this study would not be accepted.
Major 3: Abstract and introduction imply that the 'ElectroPhysiome' refers to models that incorporate both the connectome and individual neuron physiology. However, the work presented in this study does not make use of any connectomics data. To make the claim that ElectroPhysiomeGAN can jointly capture both 'network interaction and cellular dynamics', the generated models would need to be evaluated for network inputs, for example by exposing them to naturalistic stimuli of synaptic inputs. It seems likely that dynamics that are currently poorly captured, like slow ramps, or the ability of the neuron to return to its resting membrane potential, will critically affect network computations.
-
