napari-mAIcrobe¶

🚀 Installation¶

Recommended (conda):

conda create -n mAIcrobe python=3.11
conda activate mAIcrobe
conda install -c conda-forge napari pyqt
pip install napari-mAIcrobe

Development installation:

conda create -n mAIcrobe python=3.11
conda activate mAIcrobe
conda install -c conda-forge napari pyqt
git clone https://github.com/HenriquesLab/mAIcrobe.git
cd mAIcrobe
pip install -e .

Installation Guide → Complete Tutorial →

✨ Why napari-mAIcrobe?¶

• Automated Segmentation: StarDist2D, Cellpose, and custom U-Net models
• AI Classification: 6 pre-trained CNN models for S. aureus cell cycle determination
• Morphological Analysis: Comprehensive measurements using scikit-image
• Interactive Interface: Intuitive napari-based GUI for all analysis steps

• Colocalization Analysis: Multi-channel fluorescence quantification
• Automated Reports: HTML reports with visualizations and statistics
• Data Export: CSV export for downstream statistical analysis
• Batch Processing: Analyze multiple images with consistent parameters

• Reproducible Workflows: Consistent analysis parameters across users
• Custom Models: Support for user-trained segmentation and classification models
• Open Source: Full transparency and community contributions
• Cross-Platform: Works on Windows, macOS, and Linux

🎯 Key Features¶

🎨 Cell Segmentation¶

Choose from multiple state-of-the-art segmentation approaches:

• StarDist2D: Optimized for bacterial cell morphology
• Cellpose: Versatile deep learning segmentation
• Custom U-Net: Train your own models for specific bacterial species

🧠 Single-Cell Classification¶

Pre-trained deep learning models for S. aureus: - DNA + Membrane models: Epi/SIM microscopy - DNA-only models: DAPI or similar staining - Membrane-only models: Phase contrast or membrane stains - Custom model support: Integrate your own classification models

📐 Comprehensive Morphometry¶

Comprehensive cell measurements: - Size metrics: Area, perimeter, equivalent diameter - Shape descriptors: Eccentricity, solidity, aspect ratio - Intensity measurements: Mean, median, standard deviation per channel - Spatial relationships: Distance to neighbors, clustering analysis

🔍 Interactive Filtering¶

Real-time cell selection based on: - Size constraints: Filter by area or diameter ranges - Shape criteria: Select specific morphologies - Intensity thresholds: Focus on cells with specific staining patterns - Classification confidence: Keep only high-confidence predictions

🔁 Analysis Workflow¶

📄 Single Image Analysis¶

1. Load Images: Phase contrast and/or fluorescence
2. Segment Cells: Choose segmentation algorithm and parameters
3. Analyze Cells: Extract morphological and intensity features and choose classification model
4. Filter Results: Interactive filtering of cell populations
5. Generate Report: Create comprehensive analysis report

📖 Documentation¶

📚 Explore the Docs¶

🧪 Sample Data¶

Access via napari: File → Open Sample → mAIcrobe (Phase, Membrane, DNA examples)

🤝 Community¶

• 🐛 Issues: https://github.com/HenriquesLab/mAIcrobe/issues
• 📚 napari hub: https://napari-hub.org/plugins/napari-mAIcrobe

📓 Available Jupyter Notebooks¶

Explore advanced functionality with included notebooks:

• Cell Cycle Model Training: Train custom classification models
• StarDist Segmentation: Retrain a StarDist segmentation model

🏗️ Contributing¶

We welcome contributions! Whether it's:

• Bug reports and fixes
• New segmentation algorithms
• Documentation improvements
• Additional test datasets
• New AI models for classification

Quick contributor setup:

git clone https://github.com/HenriquesLab/mAIcrobe.git
cd mAIcrobe
pip install -e .[testing]
pre-commit install

Testing:

pytest -v
pytest --cov=napari_mAIcrobe
tox

Full guide: https://github.com/HenriquesLab/mAIcrobe/blob/main/CONTRIBUTING.md

📜 License¶

Distributed under the terms of the BSD-3 license — free and open source software.

🙏 Acknowledgments¶

napari-mAIcrobe is developed in the Henriques and Pinho Labs with contributions from the napari and scientific Python communities.

Built with:

• napari — multi-dimensional image viewer
• TensorFlow — machine learning framework
• StarDist — star-convex object detection
• Cellpose — generalist cell segmentation
• scikit-image — image processing

Ready to analyze your microbial images? Get started now →