/pyGCluster

pyGCluster, a novel hierarchical clustering approach

Primary LanguagePythonMIT LicenseMIT

pyGCluster

INTRODUCTION

Copyright 2011-2017 by
D.Jaeger
J. Leufken (former Barth)
A. Niehues
C. Fufezan

pyGCluster is a clustering algorithm focusing on noise injection for subsequent cluster validation. By requesting identical cluster identity, the reproducibility of a large amount of clusters obtained with agglomerative hierarchical clustering is assessed. Furthermore, a multitude of different distance-linkage combinations are evaluated. Finally, highly reproducible clusters are grouped into communities, and graphical representation of the results as node maps and heat maps is implemented.

CITATION

Please cite us when using pyGCluster for your publications.

Jaeger, D., Barth, J., Niehues, A. & Fufezan, C. pyGCluster, a novel hierarchical clustering approach. Bioinformatics 30, 896–8 (2014).

INSTALLATION

python2.7 setup.py install

(you might need admin privileges to write in the python site-package folder)

SciPy and NumPy are required for running pyGCluster.

PARTICIPATE

If you like to participate, simply checkout the source code from our git hub at http://github.com/pyGCluster, include in your changes and submit them to us.

DOCUMENTATION

For more in depth documentation of the modules and examples, please refer to the documentation folder or http://pygcluster.github.io