Learning paths

Curated routes through the PhenoMe documentation and notebooks. Pick one path and follow it in order - jumping around tends to leave gaps.

Path A - First day

Goal

Start using PhenoMe today and understand the basic concepts. Target audience: newcomers with no prior PhenoMe experience.

1. Getting started - install, verify import phenome, CPU/GPU note.
2. Notebook 01 - Interactive quickstart - widgets, toy or your data.
3. FAQ - installation - common blockers.
4. Optional: Core concepts - understand terms like visual fingerprints, metadata, and results.

Path B - Your own dataset

Goal

Run a reliable analysis on your own images using experimental info (metadata). Target audience: scientists / analysts moving from a toy run to their own data.

1. Getting started - environment locked in.
2. Preparing your data - decide if you need labels; folder patterns vs spreadsheets.
3. Notebook 02 - Load and inspect images - quality checks.
4. Notebook 03 - Core phenotyping workflow - embeddings, properties, plots.
5. Select properties — Presets — choose a property_preset (column meanings: Understanding properties).
6. Workflows or Example notebooks - match your study design.

Path C - Reporting and communication

Goal

Communicate and share your results with collaborators. Target audience: PIs, reviewers, authors of a paper or slide deck.

1. Complete Path B through notebook 03 (or load saved results).
2. PhenoMe.generate_report - parameters, sections, and output.
3. FAQ - results.

Path D - Developer / extension author

Goal

Make safe changes to the codebase and keep API docs up to date. Target audience: library developers and people embedding PhenoMe in a larger pipeline.

1. Architecture - packages, data flow, extension points.
2. API & data formats.
3. Developer guide - docstrings, Astro build, CI.
4. Testing - ruff, mypy, interrogate.
5. Extending PhenoMe, Plugins, Select properties.
6. Notebook 05 - Plugins.
7. Notebook 06 - Custom model and preprocessing.

Path E - Interpretability and exploration

Goal

Visualize patterns, find links between features, and explain AI results. Target audience: scientists looking for biological patterns in their data.

1. Notebook 04 - Exploratory analysis.
2. Explaining AI results.
3. Glossary - terminology overview (visual fingerprints, linking AI to biology, etc.).

Doc vs notebook: where to look

Topic	Read first	Practice in
Install	Getting started	Notebook 01 (Colab cell if needed)
Metadata	Preparing your data	Notebook 02
Full pipeline	Getting started, step 3	Notebook 03
Plugins	Extending	Notebook 05
Custom Models	Extending	Notebook 06