https://github.com/greenelab/pathcore-t

Methods to build a network of pathway co-occurrence relationships out of expression signatures extracted from transcriptomic compendia.
https://github.com/greenelab/pathcore-t

analysis gene-expression methodology supplement

Last synced: 5 months ago
JSON representation

Methods to build a network of pathway co-occurrence relationships out of expression signatures extracted from transcriptomic compendia.

• Host: GitHub
• URL: https://github.com/greenelab/pathcore-t
• Owner: greenelab
• License: bsd-3-clause
• Created: 2017-05-25T13:11:29.000Z (over 8 years ago)
• Default Branch: master
• Last Pushed: 2017-10-17T15:51:12.000Z (almost 8 years ago)
• Last Synced: 2024-11-06T12:56:46.282Z (11 months ago)
• Topics: analysis, gene-expression, methodology, supplement
• Language: Python
• Homepage:
• Size: 21.5 KB
• Stars: 4
• Watchers: 4
• Forks: 3
• Open Issues: 1
• Metadata Files:
  • Readme: README.rst
  • License: LICENSE

Awesome Lists containing this project

README

          
PathCORE-T

----------

Python 3 implementation of methods described in

`Chen et al.'s 2017 PathCORE-T paper `_.

Note that this software was renamed from PathCORE to PathCORE-T in Oct 2017.

The T specifies that pathway co-occurrence relationships are identified using

features extracted from **transcriptomic** data. 

The module itself is still named `pathcore` to maintain backwards

compatibility for users of the original PathCORE software package. 

This code has been tested on Python 3.5.

The documentation for the modules in the package can be

`accessed here `_.

Installation

----------------

To install the current PyPI version (recommended), run::

    pip install PathCORE-T

For the latest GitHub version, run::

    pip install git+https://github.com/greenelab/PathCORE-T.git#egg=PathCORE-T

Examples

---------

We recommend that users of the PathCORE-T software begin by reviewing the

examples in the `PathCORE-T-analysis `_

repository. The analysis repository contains shell scripts and wrapper

analysis scripts that demonstrate how to run the methods in this package

on features constructed from a broad compendium according to the 

`workflow we describe in our paper `_.

Specifically, `this Jupyter notebook `_

is a simple example of the workflow and a great place to start.

Package contents

----------------

=====================================

feature_pathway_overrepresentation.py

=====================================

The methods in this module are used to identify the pathways

overrepresented in features extracted from a transcriptomic dataset

of genes-by-samples. Features must preserve the genes in the dataset

and assign weights to these genes based on some distribution.

[`feature_pathway_overrepresentation documentation. `_]

===========

network.py

===========

Contains the data structure ``CoNetwork`` that stores information

about the pathway co-occurrence network. The output from

a pathway enrichment analysis in ``feature_pathway_overrepresentation.py``

serves as input into the ``CoNetwork`` constructor.

[`CoNetwork documentation. `_]

============================

network_permutation_test.py

============================

The methods in this module are used to filter the constructed

co-occurence network. We implement a permutation test that evaluates

and removes edges (pathway-pathway relationships) in the network

that cannot be distinguished from a null model of random associations.

The null model is created by generating *N* permutations of the network.

[`network_permutation_test documentation. `_]

Acknowledgements

----------------

This work was supported by the Penn Institute for Bioinformatics