Pycytominer
A Python package for processing and normalizing high-dimensional morphological feature data from high-throughput cell imaging experiments.
What is Pycytominer?
Pycytominer is a Python package that processes and normalizes high-dimensional morphological feature data extracted from high-throughput cell imaging experiments. It transforms raw single-cell readouts into reproducible, analysis-ready profiles suitable for downstream machine learning and statistical analysis. The tool is specifically designed to handle data from platforms like CellProfiler and DeepProfiler.
Bioinformaticians, computational biologists, and researchers working with high-throughput cell imaging data who need to standardize and prepare morphological feature profiles for analysis.
Developers choose Pycytominer for its consistent, simple API tailored to image-based profiling workflows, its integration with popular tools like CellProfiler and CytoTable, and its focus on reproducibility in processing high-dimensional cell morphology data.
Overview
Python package for image-based profiling
Use Cases
Best For
- Processing morphological features from CellProfiler or DeepProfiler outputs
- Normalizing and standardizing high-dimensional cell imaging data
- Aggregating single-cell profiles to well or population-level summaries
- Selecting informative features for downstream machine learning analysis
- Creating consensus signatures from replicate experiments
- Preparing data for visualization tools like Morpheus heatmaps
Not Ideal For
- Projects using image analysis tools other than CellProfiler or DeepProfiler without CytoTable for data harmonization
- Real-time processing workflows where data needs to be analyzed as it's generated from microscopes
- Environments with severe memory limitations that cannot handle large single-cell SQLite files
Pros & Cons
Pros
All core functions like aggregate and normalize use a uniform interface with consistent arguments, making the workflow predictable and easy to script, as highlighted in the API section.
Defaults to CellProfiler's data structure with Metadata_ prefixes and compartment names, reducing setup time for standard experiments, as described in the CellProfiler support section.
Supports modern formats like parquet, compressed CSV, and AnnData, enabling efficient storage and integration with other bioinformatics tools, mentioned in the Frameworks section.
Seamlessly works with CytoTable for data preparation and other cytomining projects like Profiling-recipe, enhancing reproducibility and pipeline compatibility.
Cons
Users must rely on external frameworks like Profiling-recipe or CytoSnake for end-to-end workflows, as Pycytominer only provides standalone functions, adding complexity for full automation.
When using tools other than CellProfiler, features must be manually defined in functions like normalize, increasing error risk and setup time, as noted in the Handling inputs section.
Processing SQLite files is limited by available memory and CPU, and the README admits this dependency can be a bottleneck for large datasets.
Frequently Asked Questions
Related Projects
SimpleITKSimpleITK: a layer built on top of the Insight Toolkit (ITK), intended to simplify and facilitate ITK's use in rapid prototyping, education and interpreted languages.
ZeroCostDL4MicZeroCostDL4Mic: A Google Colab based no-cost toolbox to explore Deep-Learning in Microscopy
NeuriteNeural networks toolbox focused on medical image analysis
AICSImageIOImage Reading, Metadata Conversion, and Image Writing for Microscopy Images in Python
Found a gem we're missing?
Open-Awesome is built by the community, for the community. Submit a project, suggest an awesome list, or help improve the catalog on GitHub.