gbrault / sdt-python

Python library for fluorescence microscopy data anlysis

The sdt-python package

sdt-python is a collection of tools for analysis of fluorescence microscopy data.

It contains

• algorithms for localization of fluorescent features in images
• methods for evaluation of tracking data
• functions to evaluate brightness data
• as well as multi-color data
• support for automated determination and correction of chromatic aberrations
• methods for reading and writing single molecule data in various formats
• handling of ROIs (both rectangular and described by arbitrary paths)
• methods for simulation of fluorescence microscopy images
• much more.

A repository of tutorials is provided at https://github.com/schuetzgroup/sdt-python-tutorials. API documentation can be found at https://schuetzgroup.github.io/sdt-python.

If you use sdt-python in a project resulting in a scientific publication, please cite the software.

Installation

Using anaconda (recommended)

Convert a miniconda installation to conda forge

The following will most likely fail on a full Anaconda install, hence it is recommended to use miniconda (minimal Anaconda) First, install miniconda (Python 3.x version). Then open an Anaconda prompt and type

conda config --add channels conda-forge
conda config --set channel_priority strict
conda update --all
conda install sdt-python
conda install opencv trackpy lmfit ipympl scikit-learn pyqt

The last line installs optional, recommended packages.

Instead of converting the whole installation to conda-forge, it is possible to

Create a new environment using conda forge

This method works for Anaconda / miniconda installs.

conda create -n sdt_env -c conda-forge --strict-channel-priority sdt-python
conda install -n sdt_env -c conda-forge --strict-channel-priority opencv trackpy lmfit ipympl scikit-learn
conda activate sdt_env

The second line installs optional, recommended packages. sdt_env is the name of the new environment. For more information on conda environments, have a look here.

Using pip (untested)

Install some Python distribution. Download this source code, change into the root folder (where this README is located) and run:

python -m pip install .

Updating

If the conda installation was converted to conda forge, type

conda update sdt-python

in an Anaconda prompt.

If a separate environment is used, type

conda activate sdt_env
conda update -c conda-forge --strict-channel-priority sdt-python

If you chose an environment name different from sdt_env when installing, adapt accordingly.

Requirements

• Python >= 3.7
• matplotlib
• numpy >= 1.10
• pandas
• pims >= 0.3.0
• tifffile >= 0.7.0
• pyyaml
• pywavelets >= 0.3.0

Recommended packages

• PyQt5 >= 5.12
• opencv
• trackpy
• lmfit
• ipympl
• scikit-learn