Filtering is usually a necessary step for isolating single neurons from continuously-recorded electrophysiology data. Unfortunately, many types of filtering distort the shape of the spike waveform. We employ a simple wavelet-based technique that reduces distortion and improves cluster separation (see Wiltschko et al. 2008 ). The code is available as a MATLAB script (requiring the Wavelet toolbox) or a Python script (requiring the PyWavelets and NumPy toolboxes).
Some journals no longer allow supplemental data online, so we host it here instead. For a video (big! 230MB .avi) of transient dyskinesias induced by selective local suppression of striatal fast-spiking interneurons (Gittis et al. 2011), click here. This mouse received 1mM IEM-1460 into lateral striatum.