Stochastic dynamics of neural synapses in health and disease

Neurons communicate through chemical synapses, which release neurotransmitter molecules in response to presynaptic action potentials.  Synapses are temporarily weakened by high frequency sequences of action potentials due to a depletion of neurotransmitter resources.   This effect, known as short term depression, modulates the transfer of signals between neurons.  Most theoretical studies of short term depression use a deterministic mean-field model of synaptic dynamics despite the fact that neurotransmitter release and recovery are fundamentally stochastic processes.  We use stochastic calculus, linear response techniques and Markov chain modeling to examine the impact of stochasticity on the filtering properties of synapses and find that it fundamentally alters the way in which a synapse filters rate-coded information.  We then combine mathematical modeling with recorded data to show that the filtering properties of stochastic depressing synapses contribute to the therapeutic efficacy of deep brain stimulation as a treatment for Parkinson's disease.