SYWTDIAWP

Biology-first course

So You Want to Do Image Analysis with Python.

Excellent. This course is here to get biological researchers from “I have data and good intentions” to “I can actually run this workflow” without pretending setup, notebooks, and tooling are somehow obvious on the first try.

13 modules Beginner friendly Multi-day course GitHub Pages ready
Begin Module 1 Browse course map

Who this is for

Researchers first, programmers second

The course assumes you know your biological question and your data better than you know terminals, environments, or package managers. Each module explains why a tool matters in a lab workflow instead of treating setup as a disconnected computing exercise.

Notebook examples
Move from commands to real exploratory analysis. Microscopy workflow
Launch napari and explore a plugin-driven analysis path. Course resources
Find notebooks, supporting files, and the shortest route back in. Scientific Python tour
See which libraries people keep talking about and what they are for. Pandas and plotting
Work with tables and make plots before the image-analysis side gets heavier. Later-course analysis
Continue into segmentation, classification, and pretrained models.

Before you begin

Plan for a few short sessions and keep one terminal window open.

Recommended outcome

Finish with one working environment and one small analysis package.

Materials

Local computer access, a browser, and permission to install software.

Pacing

Built for researchers who also have experiments to run

The lessons are meant to be worked through over several days, not inhaled in one heroic sitting. Each module leaves more room for explanation, checkpoints, and sensible stopping points.

Plan your study path
Use the resources page to spread the course across several sessions. Slow down in basics
Spend extra time here before notebooks if code still feels new.

Modules

A guided path, not a software scavenger hunt

Each module includes goals, copy-paste commands, common mistakes, and a clear next step so learners do not have to infer the workflow on their own.

Module 01

Install Python locally

Choose a straightforward installation route, verify your version, and learn the few terminal actions you need to get unstuck.

Open module

Module 02

Virtual environments with venv and conda

Learn why environments matter for reproducibility and compare the simplest standard-library route with the conda ecosystem.

Open module

Module 03

Python programming basics

Build confidence with variables, strings, numbers, lists, dictionaries, loops, and simple reusable functions.

Open module

Module 04

Scientific Python libraries

Get a practical overview of the libraries that show up again and again in scientific Python workflows.

Open module

Module 05

Run Jupyter notebooks

Install the notebook tools, open example notebooks, and try beginner exercises using tabular and image data.

Open module

Module 06

pandas and matplotlib

Work with tabular measurements, clean simple data, and make plots that help you think before you model.

Open module

Module 07

Install napari and launch mAIcrobe

Set up a viewer environment, install the plugin, and orient learners around a practical microbial image-analysis workflow.

Open module

Module 08

Create your own Python package

Use the HenriquesLab cookiecutter template to generate a clean project you can adapt for your own scripts and analysis logic.

Open module

Module 09

Segment with machine learning

Move into classical ML segmentation with normalization, clustering, IoU, feature engineering, and random forests.

Open module

Module 10

Classify single cells

Use morphology-derived measurements, normalization, and scikit-learn classifiers to infer cell states.

Open module

Module 11

Filter bad segmentations

Measure objects, inspect failures, and use classification to reduce manual cleanup.

Open module

Module 12

Understand neural networks

Build intuition with tiny NumPy examples before jumping into more opaque deep-learning workflows.

Open module

Module 13

Try pretrained models

Explore Cellpose, BioImage Model Zoo, and accessible deep-learning tools in a practical research context.

Open module