The SpyKING CIRCUS is a python code to allow fast spike sorting on multi channel recordings. It is inspired from the following publication, but not exactly similar. Performance, speed and robustness of the previously described algorithm have been improved, and details can now be found on eLife. It has been tested on datasets coming from in vitro retina with 252 an up to 4225 electrodes MEA, from in vivo hippocampus with tetrodes, in vivo and in vitro cortex data with 30 and up to 128 channels, with good results. In fact, synthetic tests on all those data show that cells firing at more than 0.5Hz can be detected, and their spikes recovered with error rates at around 1%, even resolving overlapping spikes and synchronous firing. It seems to be compatible with optogenetic stimulation, based on experimental data obtained in the retina.

The documentation is now online at spyking-circus.rtfd.org, and a Google Group to discuss/comment the software is available here. The code is now avalaible as an open access tool hosted on Github.

The datasets used in the paper, with ground-truth recordings performed in mice retina are freely available on Zenodo or in the following page.