nscherf / SVF

SVF

This repository contains the Statistical Vector Flow (SVF) framework and the tissue propagation software proposed in the article In toto imaging and reconstruction of post-implantation mouse development at the single-cell level.

Description of the repository

Folders:

• IO: The class SpatialImage, a container for images and input/output. When the right external libraries are installed (see bellow), can read tiff, hdf5, klb and inr images.
• TGMMlibraries: The class lineageTree, a container for lineage trees and Statistical Vector Flow (SVF). Can read output data from TGMM.
• csv-parameter-files: Example of parameterization csv files for each algorithms. Python files:
• SVF-prop.py: python script to build Statistical Vector Flow from a TGMM dataset.
• tissue-bw-prop.py: python script to propagate tissue information from a manually annotated 3D image.

Basic usage

Each of the python scripts proposed here can be run from a terminal in the following way:

python SVF-prop.py

python tissue-bw-prop.py

The user is then prompted to provide a parameter csv file (examples provided in the folder csv-parameter-files). The path to the parameter file should then be typed in the terminal.

Dependencies

Some dependecies are requiered:

• general python dependecies:
  • numpy, scipy, pandas
• SVF-prop.py:
  • TGMMlibraries has to be installed (see TGMMlibraries README.md)
• tissue-bw-prop.py:
  • TGMMlibraries has to be installed (see TGMMlibraries README.md)
  • IO library has to be installed (see IO README.md)

Quick install

To quickly install the script so it can be call from the terminal and install too the common dependecies one can run

python setup.py install [--user]

Still will be remaining to install IO and TGMMlibraries packages.