Visualization for AIND ephys pipeline

aind-ephys-visualization

Description

This capsule is designed to visualize ephys and spike sorted data for the AIND pipeline.

Visualizations is done using the sortingview backend of the spikeinterface.widgets module and uses the Figurl technology to produce cloud-based shareable links.

Two types of visualizations are produced:

• traces: raw, preprocessed, and drift visualizations (see example)
• sorting summary: spike sorting results for visualization and curation (see example)

Inputs

The data/ folder must include:

• the original session data (e.g., "ecephys_664438_2023-04-12_14-59-51")
• the output of the aind-ephys-preprocessing capsule
• the output of the spike sorting capsule (either with aind-ephys-spikesort-pykilosort or aind-ephys-spikesort-kilosort25)
• the output of the aind-ephys-postprocessing capsule
• the output of the aind-ephys-curation capsule (optional)
• the output of the aind-ephys-unit-classifier capsule (optional)

Parameters

The code/run script takes no arguments.

A list of visualization parameters can in the code/params.json:

{
    "job_kwargs": {
        "chunk_duration": "1s",
        "progress_bar": false
    },
    "visualization": {
        "timeseries": {
            "n_snippets_per_segment": 2,
            "snippet_duration_s": 0.5,
            "skip": false
        },
        "drift": {
            "detection": {
                "peak_sign": "neg", 
                "detect_threshold": 5, 
                "exclude_sweep_ms": 0.1
            },
            "localization": {
                "ms_before": 0.1, 
                "ms_after": 0.3, 
                "radius_um": 100.0
            },
            "n_skip": 30,
            "alpha": 0.15,
            "vmin": -200,
            "vmax": 0,
            "cmap": "Greys_r",
            "figsize": [10, 10]
        },
        "motion": {
            "cmap": "Greys_r", 
            "scatter_decimate": 15, 
            "figsize": [15, 10]
        }
    }
}

Output

The output of this capsule is the following:

• results/visualization_{recording_name}.json, a JSON file including the visualization links
• results/data_process_postprocessing_{recording_name}.json file, a JSON file containing a DataProcess object from the aind-data-schema package.