CompNeuroVis

CompNeuroVis connects computational neuroscience simulations and data to interactive desktop visualizations. Sources expose model data and controls; views opt into line plots, morphology, surfaces, and other panels; cnv.show() integrates those pieces into one application.

Release Status

CompNeuroVis 0.4.0 is the first stable release. Until 1.0.0, public APIs may still change between minor releases; changes are recorded in the changelog.

This release supports the inline source workflow used by current examples. Notebook, distributed, and multi-source composition paths remain experimental.

Installation

CompNeuroVis requires Python 3.11.

Install the latest release:

pip install compneurovis

Install current checkout:

pip install -e .

Install optional simulator integrations:

pip install -e ".[neuron]"
pip install -e ".[jaxley]"

Notebook dependencies are experimental:

pip install -e ".[notebook]"

Minimal App

import math

import compneurovis as cnv


state = {"time_ms": 0.0, "frequency_hz": 1.0}


def step(ctx):
    state["time_ms"] += 16.0


def set_frequency(ctx, value):
    state["frequency_hz"] = float(value)


src = cnv.source(step)

wave = src.line(
    "Sine wave",
    read=lambda: math.sin(
        2.0
        * math.pi
        * state["frequency_hz"]
        * state["time_ms"]
        / 1000.0
    ),
    x=lambda: state["time_ms"],
    y_min=-1.1,
    y_max=1.1,
)

src.slider(
    "frequency_hz",
    label="Frequency (Hz)",
    get=lambda: state["frequency_hz"],
    set=set_frequency,
    min=0.1,
    max=5.0,
)

cnv.layout(((wave,), (src.controls_panel,)))
cnv.show(title="Sine wave")

Run repository version:

python examples/custom/sine_wave.py

NEURON Morphology

NEURON sources keep simulator-specific data collection optimized while using the same line, control, layout, and show API.

from neuron import h

import compneurovis as cnv


soma = h.Section(name="soma")
soma.L = soma.diam = 20.0
soma.insert("hh")

stim = h.IClamp(soma(0.5))
stim.delay = 100.0
stim.dur = 5.0
stim.amp = 0.8


def set_stimulus(ctx, value):
    stim.amp = float(value)


src = cnv.neuron.source(
    sections=[soma],
    dt=0.025,
    display_dt=0.5,
)

morphology = src.morphology(
    variable="v",
    name="Membrane voltage",
    unit="mV",
    color_limits=(-80.0, 50.0),
    selected="soma@0.50000",
)

voltage = src.line(
    "Selected voltage",
    source=morphology.selection,
    y_unit="mV",
    rolling_window=500.0,
)

src.slider(
    "stimulus",
    label="Stimulus amplitude (nA)",
    get=lambda: stim.amp,
    set=set_stimulus,
    min=0.0,
    max=2.0,
)

cnv.layout(((morphology, voltage), (src.controls_panel,)))
cnv.show(title="NEURON morphology")

Full showcase:

python examples/neuron/complete_interface.py

Static surface showcase:

python examples/widgets/surface.py

Supported Surface

• cnv.source(...) for Python callables, iterators, and static data.
• cnv.neuron.source(...) and cnv.jaxley.source(...) for simulator-native sources.
• Source-level line, bar, morphology, and surface views.
• Typed controls, buttons, and hotkeys used by current examples.
• cnv.layout(...) for single-app panel placement.
• cnv.show(...) with VisPy and PyQt6 desktop rendering.

Views are opt-in. A simulator source can own morphology without displaying a morphology panel.

Experimental

These paths are incomplete or not release-supported:

• Notebook frontends and notebook process lifecycle.
• Remote sources and remote actors.
• Source composition and independent multi-source execution.
• Multi-source custom layouts.
• Advanced callback access through backend internals.

Incomplete distributed entrypoints live under cnv.experimental so they are not mistaken for supported root-level API.

Documentation

Start with Getting Started, then follow the ordered Example Path. To add an app-local or distributable widget, use Third-Party Widget Authoring.

Build the documentation site locally:

python -m mkdocs build --strict

License

Apache License 2.0.