Correlogram view in development version of phy (Unexpected behavior)

[Shavika Rastogi]

Hello everyone !

I am new to spike sorting and I am using klustakwik to do spike sorting on my datasets. I am using latest development version of phy for visualisation. As per my knowledge, refractory period should be clearly visible in the autocorrelogram and there should not be any spike in this period. I have a number of datasets and for each dataset, I can observe peak in the middle in autocorrelogram as shown in the attached figure.

Could anyone explain me the reason behind this?  Or is it because development version of phy interprets correlogram in different way?

 

Hoping for an early response.

With Warm Regards

Shavika Rastogi

M.Sc. (Neuroscience)

University of Freiburg, Germany

[Cyrille Rossant]

When it comes to the very central peak of the CCG, it is true that the development version of phy shows it, whereas earlier versions did not. This should be the only change in the CCG computation algorithm.

In any case, spikes occurring within the refractory period might indicate failures of the automatic spike sorting algorithm, which can be tackled during the manual stage with a GUI like phy.

Cyrille

--
You received this message because you are subscribed to the Google Groups "phy-users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to phy-users+...@googlegroups.com.
To view this discussion on the web, visit https://groups.google.com/d/msgid/phy-users/c051565d-b71a-4f1e-96d8-180c2e85200c%40googlegroups.com.