Fig. 6
From: Inferring and validating mechanistic models of neural microcircuits based on spike-train data
Estimation results for synaptic couplings using in vivo data. a Example spike trains of a juxtacellularly recorded PYR and an extracellularly recorded INT, with stimulation pulses indicated (top panel). Instantaneous spike rate of the INT computed via kernel density estimation using a Gaussian kernel with width σG as indicated (bottom panel). b CCG between spontaneous PYR and INT spike times and as computed from an I&F neuron fitted to the INT spike train with estimated coupling strength and delay, corresponding to the example in (a). Note that synaptic coupling was modeled using delayed, transient pulses. Method 1a was used for parameter estimation. c ROC curves across 78 PYR-INT pairs using the spontaneous PYR spikes (up to 1000 per pair) for three different values of σG. Maximal (balanced) accuracy values are indicated
