***********************************************************
CSM for bingo2 - 2010
***********************************************************

Contact: laurent.boyer@univ-st-etienne.fr

This program is a command line tool to discover Transcription Factor Binding Site in sequences of orthologous
pairs of promoters. This is part of ANR Project BINGO2 (http://bingo2.greyc.fr/).
All the parts are written to be used on LINUX.

Authors : Laurent Boyer, Elisa Fromont, Olivier Gandrillon, Amaury Habrard, Mathilde Pellerin and Marc Sebban

TOC:

1 Introduction
2 Requirement
3 Data
4 Scripts
5 Examples
6 Notes


***********************************************************
1 Introduction
***********************************************************

Some details can be found in the paper:

Laurent Boyer, Olivier Gandrillon, Amaury Habrard, Mathilde Pellerin, Marc Sebban, 
Learning Constrained Edit State Machines, 
Tools with Artificial Intelligence, IEEE International Conference on, pp. 734-741, 2009 21st IEEE International
Conference on Tools with Artificial Intelligence, 2009.


***********************************************************
2 Requirement
***********************************************************

Java 1.5 (or more recent versions)


***********************************************************
3 Data
***********************************************************

- model.init
structure of model to learn


Some data used to learn a model for the transcription factor named ZF5_B.

- ZF5_B/init
learning sample labelled by Match(TM) 

- ZF5_B/sample
learning sample without labelling

- ZF5_B/eva
pairs of orthologous TFBS


Some test files.

- ZF5_B/pairs
example of pairs of orthologous promoters

- ZF5_B/singles
example of single promoter

- ZF5_B/match_result.txt
results of Match(TM) on some promoters


***********************************************************
4 Scripts
***********************************************************

A set of scripts are provided:

- plotProm.plot
 
- visu.sh

- learn.sh

- test_pairs.sh

- test_single.sh


***********************************************************
5 Examples
***********************************************************

In order to learn the model use the script learn.sh. This script create a file named model.learned.

The scripts test*.sh are used in order to find tfbs on promoters. 

test_pairs.sh works on pairs on promoters and create a file pairs.tfbs. In this file, you have the retrieved tfbs
and their position. 

test_single.sh handles single promoter. In order to visualise the result use the script visu.sh. This one creates 
a graph which shows the probability for each position of the promoter to be in the core of the tfbs.


***********************************************************
6 Notes
***********************************************************

Note that this works is still processing. Changes must occur without notice.


***********************************************************
The end
***********************************************************