Flexible statistical inference for mechanistic models of neural dynamics

TL;DR

This paper introduces a neural network-based likelihood-free inference method for Bayesian parameter estimation in complex single-neuron models, enabling accurate data fitting without model-specific algorithms.

Contribution

It presents a novel ABC approach using neural networks to efficiently estimate posterior distributions in mechanistic neural models, handling missing data and feature selection automatically.

Findings

Accurately recovers ground-truth parameters on synthetic data.

Matches empirical voltage traces in in-vitro recordings.

Efficiently estimates multivariate posteriors over biophysical parameters.

Abstract

Mechanistic models of single-neuron dynamics have been extensively studied in computational neuroscience. However, identifying which models can quantitatively reproduce empirically measured data has been challenging. We propose to overcome this limitation by using likelihood-free inference approaches (also known as Approximate Bayesian Computation, ABC) to perform full Bayesian inference on single-neuron models. Our approach builds on recent advances in ABC by learning a neural network which maps features of the observed data to the posterior distribution over parameters. We learn a Bayesian mixture-density network approximating the posterior over multiple rounds of adaptively chosen simulations. Furthermore, we propose an efficient approach for handling missing features and parameter settings for which the simulator fails, as well as a strategy for automatically learning relevant features using recurrent neural networks. On synthetic data, our approach efficiently estimates posterior distributions and recovers ground-truth parameters. On in-vitro recordings of membrane voltages, we recover multivariate posteriors over biophysical parameters, which yield model-predicted voltage traces that accurately match empirical data. Our approach will enable neuroscientists to perform Bayesian inference on complex neuron models without having to design model-specific algorithms, closing the gap between mechanistic and statistical approaches to single-neuron modelling.