neuropixels-analysis

Analyze Neuropixels electrophysiology data with Kilosort4 and export curated units.

4|Updated Mar 2, 2026
One-click install
npx skills add https://github.com/shushuzn/Rairos --skill neuropixels-analysis-shushuzn
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: neuropixels-analysis
Source: https://github.com/shushuzn/Rairos/tree/main/skills/neuropixels-analysis
Command: npx skills add https://github.com/shushuzn/Rairos --skill neuropixels-analysis-shushuzn

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires spikeinterface, neuropixels-analysis, kilosort, spykingcircus, mountainsort5, anthropic, and includes scripts (resource) and references (resource) and assets (resource) components.

What problem does it solve?

Neuropixels recordings require a complex, error-prone workflow—from loading raw data through preprocessing, drift correction, spike sorting, QC, and unit curation—so this skill streamlines the full pipeline into reproducible steps that produce publication-ready curated results.

Core Features & Use Cases

  • End-to-end Neuropixels analysis workflow: Load SpikeGLX/Open Ephys/NWB data, preprocess, correct motion/drift, run spike sorting (including Kilosort4), compute quality metrics, and curate units using Allen/IBL-style criteria.
  • Publication-ready quality assessment & curation: Generate metrics like SNR, ISI violations, presence ratio, amplitude cutoff, drift metrics, isolation measures, and support exporting for manual verification.
  • AI-assisted visual analysis for uncertain units: Use vision-language models to review waveform/correlogram plots and make curation decisions for borderline cases.
  • Export for downstream review and sharing: Export curated results to Phy for human-in-the-loop inspection and to NWB for standardized archiving.

Quick Start

Run a full Neuropixels pipeline by loading your recording data, preprocessing it, running Kilosort4 spike sorting, computing quality metrics, and curating units with an Allen-style method in one go.

Frequently Asked Questions about neuropixels-analysis

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I run spike sorting and motion correction for Neuropixels data end-to-end?

To run spike sorting for Neuropixels data, you can load SpikeGLX, Open Ephys, or NWB files to preprocess, apply drift correction, run Kilosort4, and compute quality metrics in a single reproducible pipeline.

What quality metrics are used for spike sorting unit curation?

Unit curation relies on quality metrics like SNR, ISI violations, presence ratio, amplitude cutoff, drift metrics, and isolation measures, applying Allen or IBL-style criteria to evaluate sorted units.

Does this workflow support phase-shift handling for Neuropixels 1.0 probes?

Yes, the workflow explicitly requires phase-shift handling for Neuropixels 1.0 probes during preprocessing to ensure accurate data alignment before drift estimation and correction.

Can I use AI to review uncertain units during spike sorting curation?

Yes, AI-assisted visual analysis uses vision-language models to review waveform and correlogram plots, making curation decisions for borderline or uncertain units to streamline the process.

How do I export curated spike sorting results for downstream review?

You can export curated results to Phy for human-in-the-loop manual inspection and to NWB format for standardized archiving, facilitating downstream analysis and data sharing.

What is the best way to handle drift correction in extracellular electrophysiology data?

The best way to handle drift correction is through a pipeline that performs drift estimation and motion correction before spike sorting, ensuring motion artifacts do not degrade unit separation quality.