Abstract
This paper deals with the problem of inferring the signals and parameters that cause neural activity to occur. The ultimate challenge being to unveil brain’s connectivity, here we focus on a microscopic vision of the problem, where single neurons (potentially connected to a network of peers) are at the core of our study. The sole observation available are noisy, sampled voltage traces obtained from intracellular recordings. We design algorithms and inference methods using the tools provided by stochastic filtering that allow a probabilistic interpretation and treatment of the problem. Using particle filtering, we are able to reconstruct traces of voltages and estimate the time course of auxiliary variables. By extending the algorithm, through PMCMC methodology, we are able to estimate hidden physiological parameters as well, like intrinsic conductances or reversal potentials. Last, but not least, the method is applied to estimate synaptic conductances arriving at a target cell, thus reconstructing the synaptic excitatory/inhibitory input traces. Notably, the performance of these estimations achieve the theoretical lower bounds even in spiking regimes.
Highlights
Measurements of membrane potential traces constitute the main observable quantities to derive a biophysical neuron model
The dynamics of auxiliary variables and the model parameters are inferred from voltage traces, in a costly process that typically entails a variety of channel blocks and clamping techniques, as well as some uncertainty in the parameter values due to noise in the signal
The problem investigated in this paper considers recordings of noisy voltage traces to infer the hidden gating variables of the neuron model, as well as filtered voltage estimates, model parameters, and input synaptic conductances
Summary
Measurements of membrane potential traces constitute the main observable quantities to derive a biophysical neuron model. The problem investigated in this paper considers recordings of noisy voltage traces to infer the hidden gating variables of the neuron model, as well as filtered voltage estimates, model parameters, and input synaptic conductances. The generation of action potentials is regulated by internal drivers (e.g., the active gating variables of the neuron) as well as exogenous factors like excitatory and inhibitory synaptic conductances produced by pools of connected neurons. This system is unobservable, in the sense that we cannot. The ultimate goal is to extract the following quantities: 1. The time-evolving states characterizing the neuron dynamics, including a filtered membrane potential and the dynamics of the gating variables
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have