Application of large deviations theory to spontaneous action potential generation

Scientific context

Neurons are subject to various sources of noise and intrinsic (ion channel stochasticity). Ion channels are macromoleculesembedded within the membrane, and their voltage-induced conformational changes is subject to thermal noise. As the mechanism of action potential generation relies on opening and closing of ion channels, which enable ion transfer through the membrane, it is intrinsically stochastic. In the limit of large numbers of ion channels, the neuron behavior is almost deterministic, and fluctuations remains due to a finite-size effect. Thus, in case where a deterministic model would predict that the membrane potential should remain at its resting state, one observes, in simulations and experiments, that spontaneous action potentials occur at a low rate, because of the remaining fluctuations. Our aim is to understand in more details this phenomenon, through the application of large deviations theory.

Computational context

Issues

Plateform improvements

Scientific results