Spike Detection

Data sets of fluorescent values recorded at a rate of 0.206 s from GCaMP6F-expressing unstimulated neurons in the myenteric plexus of the proximal large intestine and analyzed with the MLspike software for Matlab downloaded from GitHub (https://github.com/MLspike/spikes) (Deneux et al., 2016).

MLspike determines a new baseline to subtract from the raw fluorescence data to allow accurate modelling. The software uses a version of the Viterbi algorithm to obtain the most probable spike train. From the model, fluctuating baseline, model-estimated spike train, and most probable spike times were extracted. Polynomial coefficients (p₂) were changed from 0.5 to 0.55 and (p₃) from 0.01 to 0.03 for GCaMP6F fluorescence as recommended (Deneux et al., 2016). The minimum range for baseline (b_min) was changed from 0.7 to 0.5 as determined by observed fluorescence values.