hc-2 data: the theta phase and cell type

Hello,

I have two questions about the hc-2 data. Firstly, since the action potentials may be detected by different electrodes on one shank, how can I get the theta phase of a neuron? In other words, given a spike train of one neuron, the LFP signal of which channel should be employed to study the phase relation between the spikes and LFP?

Secondly, the cells type can be separated by their autocorrelograms, waveforms and mean firing rates, but how can I get the spike waveform? By using the matlab script “Spk = LoadSpk('ec013.527.spk.1',8)”, for example, I got a 3d array Spk(Channel, Sample, Spike Number) array of 8*32*39659 elements. According to the “ec013.527.clu.1” file, the 966th spike is classified as cluster 12, but there are eight waveforms from eight electrodes, which waveform should be used? Whether the citation of largest amplitude can be employed to identify the correspondence? Additionally, in the description of the hc-2 data it states that the width of each spike is 32 samples with sample #15 being the trough of spike. But I found this does not always hold, why? In the text file of waveshape features for each spike (‘ec013.527.fet.1’ for example), there are 29 features, which is the spike width? One more thing, what is the appropriate epoch duration can be used to calculate the mean firing rate?

Thank you!