Neural signals / Software & preprint
Spike Sorting Framework
From voltage to distinct waveforms
From neural recordings to interpretable spike groups: PyWaveClus and MCWs.
01 / Inside the method
How it works.
Select a stage to explore the workflow.
Conceptual schematic · Simplified method overview
Electrophysiological recordings contain neuronal activity and noise. The recording environment shapes the preparation and quality checks needed.
Candidate spikes are detected and short waveform segments are extracted for comparison.
Waveform representations support comparisons between candidate events. PyWaveClus uses Haar wavelet and PCA features; see each project for its specific implementation.
Candidate events are organized into spike groups for analysis. PyWaveClus uses superparamagnetic clustering; MCWs targets reliable sorting in human hospital recordings.
02 / Context & sources
The research.
Two approaches to spike sorting
Spike sorting separates candidate neuronal events in electrophysiological recordings into groups with similar characteristics. My work in this area includes PyWaveClus, a Python pipeline, and MCWs (MiCroWire sorter), a collaborative framework for human intracerebral recordings.
The schematic above shows the shared task at a conceptual level. The two projects have their own methods and implementations.
PyWaveClus
PyWaveClus implements a Python workflow inspired by Wave_clus. It brings together spike detection, waveform extraction, Haar wavelet and PCA features, and superparamagnetic clustering.
Its documentation covers artifact removal and the steps required to run the pipeline on electrophysiological recordings, with credit to the original Wave_clus and superparamagnetic clustering methods. Sorting output requires quality assessment for the recording being studied.
Explore PyWaveClus code and documentation
MCWs · MiCroWire sorter
MCWs is a framework for automated spike sorting in human intracerebral recordings. It addresses the recording-quality challenges of hospital environments, including noise and interruptions during experimental sessions.
Developed with colleagues in ReyLab, this work supports analysis of human neuronal activity. I am a coauthor of the framework’s preprint.
Alexander Betancourt, Masoud Khani, Tapasi Brahma, Fernando J. Chaure, Connor Hauder, Sunil Mathew, Ana Sofia Dominguez Zesati, Sean Lew, Kunal Gupta, and Hernan G. Rey. MCWs (MiCroWire sorter): A new framework for automated and reliable spike sorting in human intracerebral recordings. bioRxiv, 2025.
The linked MCWs work is a preprint and has not been peer reviewed.