Personal tools
You are here: Home Events ANC Workshop: Paolo Puggioni and Amos Storkey, Chair: Charles Sutton

ANC Workshop: Paolo Puggioni and Amos Storkey, Chair: Charles Sutton

— filed under:

What
  • ANC Workshop Talk
When Mar 25, 2014
from 11:00 AM to 12:00 PM
Where IF 4.31/4.33
Add event to calendar vCal
iCal

Paolo Puggioni

Bayesian inference of synaptic inputs given an in vivo voltage clamp trace

Voltage clamp recordings in awake animals directly measure the excitatory (or inhibitory) synaptic current flowing in a single neuron.

The current results from the summation of thousands of synaptic inputs of variable size per second. Recent methods based on Kalman filtering [Paninski2012, Lankarany2013] give a good estimation of the time-varying synaptic conductance, that is a function of input rate and size. However, inferring how the distribution of input size changes during behaviour can be important to understand computational mechanisms used by our brain. Here I propose a simple hierarchical model that is able to estimate both frequency and probability distribution of the synaptic weights. Firstly I test the validity of my model by using generated data, showing that it is robust to high frequency noise and slow fluctuations in the inputs rate typical of in vivo recordings. Thus, I apply it to voltage clamp traces recorded from interneurons in the cerebellum of awake mice.