FuNeL - Functional Networks Learning co-prediction protocol.

Copyright (C) 2014-2015  Nicola Lazzarini
School of Computer Science, Newcastle University

This program is free software: you can redistribute it and/or modify it
under the terms of the GNU General Public License as published by
the Free Software Foundation, either version 3 of the License, or
(at your option) any later version.

This program is distributed in the hope that it will be useful, but
WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for
more details: http://www.gnu.org/licenses/gpl.html


DESCRIPTION
-----------

FuNeL is a protocol to infer functional networks from machine learning models.
It is a general approach that uses BioHEL, a rule-based evolutionary classifier,
to describe the expression samples as a set of rules and then infers interactions
between genes that act together within rules to predict the samples class.

FuNeL generates **co-prediction** networks, that capture biological knowledge
complementary to that contained in popular similarity-based co-expression networks.


REQUIREMENTS
------------

FuNeL is a set of scripts written in several programming/scripting languages
and is able to run on GNU/Linux or MacOS systems.

To run it requires the following software installed:

* Weka - http://www.cs.waikato.ac.nz/ml/weka/ (GNU General Public Licence)
* Python 2.7.x - https://www.python.org/ (Python Software Foundation Licence)
* NumPy - http://www.numpy.org/ (BSD Licence)
* GNU R - http://www.r-project.org/ (GNU General Public Licence)


COMPILATION
------------

To run FuNeL, you need to compile BioHEL first. Its source code is distributed
along with FuNeL scripts. Use the Makefile provided in BioHEL folder:

	cd BioHEL
	make


CONFIGURATION
--------------

Inside coprediction.sh there are the following variables:

* WEKA_PATH - path to the weka.jar file from the Weka programme
* NUM_BIOHEL_RUNS - number of BioHEL runs
* NUM_PERMUTATIONS - number of permuted datasets used for significance testing


RUN
----

To perform co-prediction with FuNeL run the following script from the main directory:

	./coprediction.sh <project_name> <dataset> <configuration> [<attributes>]

* project name - name of the results directory
* dataset - gene expression data in ARFF format
* configuration - protocol variant (number from 1 to 4) used to generate the network
* attributes - number of attributes to retain from feature selection (only for configurations 1 and 3)

It will create the resulting `co-prediction_network.txt` and other intermediate
files in the `results/<project_name>` directory.