zeztz-flow

zeztz-flow is an marimo-compatible flow-cytometry FCS analysis package. Derived from cytoflow, this package is designed to become a minimal platform, allowing users to perform vibe analysis with their own AI agents.

Features:

• FCS import with automatic well-plate metadata
• Cleanup gating: scatter, doublet removal, polygon, custom thresholds
• Density plots, faceted histograms, ridgeline plots
• Subset by row, column, or well
• Median-fluorescence export
• Self-contained — vendored Cytoflow runtime, no extra installs
• Pure Python — no compiled extensions
• Designed for AI-assisted analysis

Installation

pip install zeztz-flow

Package name on PyPI is zeztz-flow (with hyphen), but the import is zeztzflow (no hyphen, no underscore):

import zeztzflow as flow

Using with marimo (uv-based)

# Install into existing marimo environment
uv pip install zeztz-flow

# Or create a new project
uv init my-fcs-analysis
cd my-fcs-analysis
uv add marimo zeztz-flow
uv run marimo edit

Development

git clone https://github.com/anomalyco/ZeztzFlow.git
cd ZeztzFlow
uv venv
uv pip install -e .
uv run python test/package_smoke_test.py

For conda-based setup: mamba env create -f environment.yml

Quick Start

Open a marimo notebook and paste this prompt to your AI agent:

Run a first-pass flow cytometry analysis using zeztzflow and marimo on the FCS files in path/to/fcs_directory. Before writing any code, find and read the docs of the installed packages — there should be a template for zeztzflow as well.

The agent reads AGENTS.md and _marimo_template.py from the installed package to follow the default pipeline, channel conventions, and marimo gotchas.

Documentation

• PROJECT_GUIDE.md — API reference, workflows, testing, publishing
• AGENTS.md — guide for AI coding agents
• llms.txt — machine-readable project map
• examples/marimo_template.py — canonical marimo notebook template

Testing

MPLCONFIGDIR=/tmp/mplconfig python test/package_smoke_test.py

License

GPL-compatible. See LICENSE.txt and THIRD_PARTY_NOTICES.md.