Metadata-Version: 2.5
Name: neuroseq
Version: 0.31.0
Summary: Qt-based desktop GUI application for computational neuroscience, combining Brian2 spiking neural network simulation with interactive stimulus design, multi-source experimental data import, and hardware export for holographic stimulation (HOLO4D .seq files).
Project-URL: Homepage, https://neuroseq.github.io/
Project-URL: Documentation, https://neuroseq.github.io/
Project-URL: Source, https://github.com/neuroseq/neuroseq
Project-URL: Issues, https://github.com/neuroseq/neuroseq/issues
Author: Andrey Formozov
License-Expression: MIT
License-File: LICENSE
Keywords: brian2,calcium imaging,gui,holographic stimulation,neuroscience,optogenetics,spiking neural networks
Classifier: Environment :: X11 Applications :: Qt
Classifier: Intended Audience :: Science/Research
Classifier: Operating System :: MacOS
Classifier: Operating System :: Microsoft :: Windows
Classifier: Operating System :: POSIX :: Linux
Classifier: Programming Language :: Python :: 3
Classifier: Programming Language :: Python :: 3 :: Only
Classifier: Programming Language :: Python :: 3.12
Classifier: Programming Language :: Python :: 3.13
Classifier: Programming Language :: Python :: 3.14
Classifier: Topic :: Scientific/Engineering
Classifier: Topic :: Scientific/Engineering :: Visualization
Requires-Python: >=3.12
Requires-Dist: brian2
Requires-Dist: cellpose[gui]
Requires-Dist: click
Requires-Dist: fastplotlib
Requires-Dist: h5py
Requires-Dist: igor2
Requires-Dist: matplotlib
Requires-Dist: numpy
Requires-Dist: nvidia-ml-py
Requires-Dist: pillow
Requires-Dist: platformdirs
Requires-Dist: psutil
Requires-Dist: pyparsing!=3.2.4
Requires-Dist: pyqt6
Requires-Dist: pyqt6-qt6!=6.11.1
Requires-Dist: pyqtgraph!=0.14.0
Requires-Dist: qtawesome
Requires-Dist: qtpy
Requires-Dist: scipy
Requires-Dist: suite2p>=1.0
Requires-Dist: tifffile
Requires-Dist: torch
Requires-Dist: torchvision
Provides-Extra: chat
Requires-Dist: anthropic>=0.40; extra == 'chat'
Requires-Dist: httpx>=0.27; extra == 'chat'
Requires-Dist: keyring>=24; extra == 'chat'
Requires-Dist: openai>=1.40; extra == 'chat'
Description-Content-Type: text/markdown

# NeuroSeq: Neuronal Sequencer

[![CI](https://github.com/neuroseq/neuroseq/actions/workflows/ci.yml/badge.svg)](https://github.com/neuroseq/neuroseq/actions/workflows/ci.yml)
[![PyPI](https://img.shields.io/pypi/v/neuroseq)](https://pypi.org/project/neuroseq/)
[![Docs](https://img.shields.io/badge/docs-neuroseq.github.io-blue)](https://neuroseq.github.io/)

NeuroSeq controls the input to a neural network and reads out its output, whether the network is simulated, biological or neuromorphic. The same input protocol and the same analysis run on all three.

| Network      | Input                                                             | Readout                                     |
| ------------ | ----------------------------------------------------------------- | ------------------------------------------- |
| Digital      | Brian2 (built in) or external simulators                          | Voltage traces, spikes, synaptic weights    |
| Biological   | Two-photon holographic optogenetics (HOLO4D/SysCon2), patch clamp | Calcium imaging (ScanImage, suite2p), Igor2 |
| Neuromorphic | BrainScaleS-2 spike sources                                       | Spikes and membrane traces                  |

![NeuroSeq GUI](https://raw.githubusercontent.com/neuroseq/neuroseq/main/docs/images/screenshot.png)

## Why

Simulations expose every input and every variable. Biological and neuromorphic networks usually do not, so the three fields developed separate methods. Modern all-optical and electrophysiological interfaces now give cell-level control and readout of living networks. NeuroSeq uses this to treat every network as a programmable system: design an input, deliver it, read the response, decide the next input.

This lets you:

- run real experiments and neuromorphic hardware almost like simulations,
- run models as if they were experiments, with realistic instrument constraints,
- test a protocol in simulation, then run it unchanged on a microscope or chip.

## Features

- **Input design**: draw, paste or script input arrays (static input), or run interactive routines that choose the next input from the last response (dynamic input, e.g. rheobase search, STDP tuning).
- **Neuron ID matching**: links input rows to real cells via Cellpose segmentation.
- **Delivery**: Brian2, external simulators, SysCon2 `.seq` scripts, ScanImage acquisition, BrainScaleS-2, patch clamp input export.
- **Readout and analysis**: import imaging (Cellpose + suite2p), Igor2 and patch clamp data; compare with simulations; extend with analysis plugins.
- **Embedded Python console** and an experimental LLM assistant.

## Install

Python 3.12 or newer:

```bash
pip install neuroseq
neuroseq
```

See the [installation guide](https://neuroseq.github.io/getting_started/installation/) (GPU, Windows) and the [developer guide](https://neuroseq.github.io/developer/developer_installation/) (uv or Conda checkout).

## Documentation

[neuroseq.github.io](https://neuroseq.github.io/). Contributions to the docs are welcome, see the [documentation guide](https://neuroseq.github.io/developer/documentation/).

## Contributing

Extensions go in as plugins where possible: new network backends, analysis methods and input protocols. They need no core changes and get a lighter review. Changes to the core architecture should be discussed with the core developers first.

- Bugs and feature requests: [GitHub issues](https://github.com/neuroseq/neuroseq/issues)
- Code: fork, branch, open a pull request. See the [contributing guide](https://neuroseq.github.io/developer/contributing/).

## License

[MIT](https://github.com/neuroseq/neuroseq/blob/main/LICENSE). Copyright © 2024-2026 Andrey Formozov. Contributors are listed in [AUTHORS.md](https://github.com/neuroseq/neuroseq/blob/main/AUTHORS.md).
