Python module to read BGEN files.


Keywords
bioinformatics, format, BGEN, binary, genetics, genomics
License
MIT
Install
pip install pybgen==0.7.1

Documentation

pybgen - Module to process BGEN files

Build Status Coverage Status PyPI version

PyBGEN is a Python module to read and write BGEN binary files and extract dosage data.

A short documentation is available at https://lemieuxl.github.io/pybgen/.

Dependencies

The tool requires a standard Python installation (2.7 and 3.6 or higher are supported) with the following modules:

  1. numpy version 1.12.0 or latest
  2. six version 1.10.0 or latest

The tool has been tested on Linux, but should work on MacOS and Windows operating systems as well.

Installation

Using pip:

pip install pybgen

Using conda:

conda install pybgen -c http://statgen.org/wp-content/uploads/Softwares/pybgen

It is possible to add the channel to conda's configuration, so that the -c http://statgen.org/... can be omitted to update or install the package. To add the channel, perform the following command:

conda config --add channels http://statgen.org/wp-content/uploads/Softwares/pybgen

Updating

To update the module using pip:

pip install -U pybgen

To update the module using conda:

# If the channel has been configured (see above)
conda update pybgen

# Otherwise
conda update pybgen -c http://statgen.org/wp-content/uploads/Softwares/pybgen

Testing

To test the module, just perform the following command:

$ python -m pybgen.tests
......................................................................
......................................................................
......................................................................
......................................................................
......................................................................
......................................................................
......................................................................
......................................
----------------------------------------------------------------------
Ran 528 tests in 13.171s

OK