leriomaggio / lego Goto Github PK

Download link:
http://omics.fudan.edu.cn/static/softwares/

README for LEGO (V2.0)
Time-stamp: <2015-10-13 12:00 by Xinran Dong & Yun Hao & Zhu Liu>

* INTRODUCTION
----------------------------------------------------------------------

We have developed a novel method named LEGO (functional Link Enrichment of Gene Ontology or gene sets). Incorporating a network-based gene-weighting scheme, LEGO measures the overlaps between the interesting genes and a given gene set by integrating both the gene weights and gene-gene interactions.   

Please Cite:

LEGO: a novel method for gene set over-representation analysis by incorporating network-based gene weights. 

* HOWTO COMPILE THE SOURCE AND INSTALL	
----------------------------------------------------------------------

	You will need a Unix environment with a working C++ compiler to
	compile the source files in src/. We recommend GCC>=4.2.
	Besides, Perl and R are necessary for running this program.

	Type
	$ perl compile.pl
	This will create a binary directory ( /bin) in the main directory.

* USAGE
----------------------------------------------------------------------

In the main directory, you could type:
	$ perl LEGO.pl
and you will see the help message.

The usage for this program: 

###############################################################################################
	perl LEGO.pl <network file> <geneset file> <interest file> [options]

	Options:  

	-h/-help: show help message.
	-pre: <whether or not to pre-run to generate mid files,1-yes,0-no(e.g:0)> 
	-multi: <multi or not:1-yes,0-no(e.g:0)> 
	-bgNE: <bgNE for the network,e.g:0.25> 
	-min: <minimum gene set size,e.g:5> 
	-max: <maximum gene set size,e.g:10000> 
	-adj: <adjusted methods,e.g:fdr> 
	-p_thre: <p value cutoff,e.g:0.05> 
	-fisher: <whether or not run fisher exact test, 1-yes,0-no,e.g: 0> 
	-filter: <whether or not run result cluster and filter step, 1-yes, 0-no, e.g: 1> 
	-bg_file: <background file, e.g: bg.txt, if do not have background file, leave it blank>

###############################################################################################

And we also provide a version that do not use permutation strategy:

The usage for this program:

###############################################################################################

perl LEGO_noperm.pl <network file> <geneset file> <interest file> [options]

        Options:

        -h/-help: show help message.
        -pre: <whether or not to pre-run to generate mid files,1-yes,0-no(e.g:0)>
        -multi: <multi or not:1-yes,0-no(e.g:0)>
        -bgNE: <bgNE for the network,e.g:0.25>
        -min: <minimum gene set size,e.g:5>
        -max: <maximum gene set size,e.g:10000>
        -adj: <adjusted methods,e.g:fdr>
        -p_thre: <p value cutoff,e.g:0.05>
        -fisher: <whether or not run fisher exact test, 1-yes,0-no,e.g: 0>
        -filter: <whether or not run result cluster and filter step, 1-yes, 0-no, e.g: 0>
        -bg_file: <background file, e.g: bg.txt, if do not have background file, leave it blank> \n";

###############################################################################################

	
Example1 (for single interesting gene list file): 
perl LEGO.pl demo/FC2_human.txt demo/GeneSet_human.txt demo/input.txt 
Example2 (for multiple interesting gene list file): 
perl LEGO.pl demo/FC2_human.txt demo/GeneSet_human.txt demo/input_mul.txt -multi 1
Example3 (for single interesting gene list file and filtered results): 
perl LEGO.pl demo/FC2_human.txt demo/GeneSet_human.txt demo/input.txt -filter 1 
Example4 (for multiple interesting gene list file and filtered results): 
perl LEGO.pl demo/FC2_human.txt demo/GeneSet_human.txt demo/input_mul.txt -multi 1 -filter 1
Example5 (for single interesting gene list file + bg file): 
perl LEGO.pl demo/FC2_human.txt demo/GeneSet_human.txt demo/input.txt -bg_file demo/bg.txt 

Example6 (for single interesting gene list file with background list): 
perl LEGO_noperm.pl demo/FC2_human.txt demo/GeneSet_human.txt demo/input.txt -bg_file demo/bg.txt
Example7 (for multiple interesting gene list file with background list): 
perl LEGO_noperm.pl demo/FC2_human.txt demo/GeneSet_human.txt demo/input_mul.txt -multi 1 -bg_file demo/bg.txt

Tips: 
	1. Output files: <interest file>_LEGO.txt, if you choose to filter & cluster enriched gene sets, another two files will be generated: <interest file>_LEGO_filter.txt <interest file>_LEGO_filter_cluster.txt (set -filter 1)
	2. The first time to run this program will take about several minutes to generate mid-files(according to the size of the network and genesets). If you want to re-generate the mid files, set `-pre_run to 1`; 
	If you have already generated the mid-files for network and genesets, it will only take several seconds to calculate a new input gene list (set `-pre_run to generate mid files` to 0)
	3. The first time to use LEGO.pl, it will take maybe several minutes to generate the permutation results. We provide the pre-generated mid files for data/GeneSet_human.txt. 
	4. If you want to use the background file, you could only use LEGO_noperm.pl;
	5. The gene ID in the network, gene set and input gene list must be the same ID system.
	6. More options are available at our online server <http://lego.tianlab.cn>. It will be available soon. 

## for plot
Pre-request:

	Usage: plot_LEGO.R <input interesting gene list file> <enriched output file> <output file prefix> <network file name> <network+gene set prefix,e.g: demo/GeneSet_human.txt_FC2_human> <CS or NW> <plot gene set name, can be comma saperated> 
## e.g:
	Rscript src/plot_LEGO.R  demo/input.txt demo/input.txt_id.out_LEGO.txt demo/input_REACTOME_KINESINS demo/FC2_human.txt demo/GeneSet_human.txt_FC2_human CS REACTOME_KINESINS