graphlets

Small package for performing graphlet decomposition.

Dependencies

Make sure you have the following installed on your machine.

A C++ compiler supporting C++11
scikit-learn
joblib
numpy
networkx

Installation

With all the dependencies installed you can install the package by running:

$ git clone https://github.com/KirillShmilovich/graphlets
$ cd graphlets
$ pip install -e .

(Note the -e is required to ensure orca/orca.cpp compiles properly)

Usage

The below examples shows how to compute a graphlet decomposition on a randomly generated set of points.

import graphlets
import numpy as np

# Create a randomly generaterd data set with dimensions (n_frames, n_objects, n_dims)
a = np.random.rand(1000, 100, 3)

# Instantiate a graphlet object using `a`
G = graphlets.Graphlets(a)

# Compute a graphlet decomposition, by default performing a
# node reduction outputing a vector of graphlet frequencies 
decomp = G.compute(r_cut = 0.1)

Acknowledgements

This package is shipped with the C++ code to perform graphlet decomposition available here:

ORCA (https://github.com/thocevar/orca)

Project based on the Computational Molecular Science Python Cookiecutter version 1.0.

References

[1] Pržulj N, Biological Network Comparison Using Graphlet Degree Distribution, Bioinformatics 2007, 23:e177-e183.

[2] Tomaž Hočevar, Janez Demšar, A combinatorial approach to graphlet counting, Bioinformatics, Volume 30, Issue 4, 15 February 2014, Pages 559–565