Toward direct links between model networks and experimental data

To understand the dynamics of brain networks we need to combine theoretical analyses of model networks and insights derived from model network simulations with the experimental data. This is a difficult task for several reasons including the level of model detail, the technical aspects of experimental recordings and different experimental contexts. A recently developed method allows one to estimate synaptic background activities using subthreshold membrane potential distributions derived from intracellular recordings. This method requires the removal of spikes to obtain the estimates. We remove spikes from heterogeneous, inhibitory network model outputs (in which the underlying dynamics are understood) and obtain synaptic distributions from multiple simulations. We show that the distributions reflect the known model network characteristics and change appropriately with different parameter values. This suggests that model network characteristics can be constrained by the experimental data and direct links between the dynamics of model and biological networks can be made.