Please see http://fusionscan.ewha.ac.kr/ for FusionScan manual and documentation.

November 13, 2014 · View on GitHub

====================================================================================================== Name : FusionScan Version : 1.0 Last Updated : 2014-10-25 Description : FusionScan is a accurate prediction tool for fusion genes from RNA-seq data. FusionScan has better precision and recall than other tools. Please see http://fusionscan.ewha.ac.kr/ for FusionScan manual and documentation.

DOWNLOAD GETTING THE SOFTWARE AND INDEX FILES : You can get the recent version of fusionscan from https://github.com/iamlife/FusionScan or under link.

FILE NAME			DESCRIPTION					RECOMENDED LOCATION
fusionscan			fusionscan src files				your wanted location
										tar xvfz fusionscan.tar.gz
										cd fusionscan/
hg19.2bit.gz			hg19.2bit file to run gfServer(blat)		/fusionscan/src_dir/
nib_files_hg19.tar.gz		hg19 nib files to use nibFrag			/fusionscan/src_dir/
bowtie2_index_hg19.tar.gz	bowtie2	index files for	hg19 & hg19.fa		/fusionscan/src_dir/
bowtie2_index_refMrna.tar.gz	bowtie2	index files for	refMrna	& refMrna.fa	/fusionscan/src_dir/
ssaha2_index_hg19.tar.gz	ssaha2 index files for hg19			/fusionscan/src_dir/
gene_dir.tar.gz			gene annotation	files				/fusionscan/
rmsk_dir.tar.gz			repeatMasker files				/fusionscan/

REQUIREMENT

ENVIROMENT TO RUN THE SOFTWARE
- JDK1.7 or Jre1.7, Python
: FusionScan's main process is developed with JAVA language and	whole-process is in Python script. 


REQUIRED INDEX AND SEQUENCE FILES
- bowtie2 index	files for hg19 & hg19.fa
- bowtie2 index	files for refMrna & refMrna.fa
- ssaha2 index files for hg19


REQUIRED PROGRAMS (go to SETTING)
- bowtie2: to obtain unmapped reads
- SSAHA2: to align reads to genome
- BLAT (gfServer, gfClient): to	confirm	final candidates
- nibFrag: to make pseudo fusion transcript
- bl2seq (blast): to search sequence homology
- fastq_quality_trimmer, fastx_collapser, fastx_artifacts_filter (FASTX-Toolkit): to do	pre-processes

HOW TO RUN

INSTALL

$tar -xvzf fusionscan.tar.gz$ cd fusionscan/

*** each directory includes ***

fusionscan/ : - Run_FusionScan.py : to run whole process

fusionscan/src_dir/: contains various programs which are used for sequence alignment, sequence comparison, getting sequences from genome.

- FusionScan.jar: FusionScan jar file

* Under	files should be	downloaded in src_dir/

- hg19.2bit: to	use gfServer
- nib_files_hg19/* : to	use nibFrag
- bowtie2_index_hg19/*:	to use BOWTIE2
- bowtie2_index_refMrna/*: to use BOWTIE2
- ssaha2_index_hg19/*: to use SSAHA2

* Under	binary src files which are proper to your system should	be copied in src_dir/ 
 in case of linux_ix86_64, all	binaries are included in fusionscan/src_dir/	

- nibFrag: from	BLAT
- gfClient: from BLAT
- gfServer: from BLAT
- bl2seq: from BLAST
- bowtie2: from	BOWTIE2
- bowtie2-align: from BOWTIE2
- bowtie2-align-debug: from BOWTIE2
- fastq_quality_trimmer: from FASTX-Toolkit
- fastx_collapser: from	FASTX-Toolkit
- fastx_artifacts_filter: from FASTX-Toolkit
- ssaha2: from SSAHA2
- ssaha2Build: from SSAHA2

fusionscan/gene_dir/ : contains various gene annotation files. These files can be downloaded from here and extract it in fusionscan/
- duplicated_gene_database_dgd_Hsa_all_v68.tsv: duplicated gene database file
- pseudogeneHUGO.txt: HGUO file's pseudogene part info file
- refGene_and_indexes.txt: indexed refGene file
- refGene.txt: raw refGene file
fusionscan/rmsk_dir/ : contains simple formatted repeatMasker data. These files can be downloaded from here and extract it in fusionscan/

SETTING 0. download index files and fasta files for hg19 and refMrna from here and extract it in fusionscan/

*** FusionScan needs bowtie2, SSAHA2, bl2seq and FASTX-Toolkit binary files.

1. Bowtie2 :	 http://sourceforge.net/projects/bowtie-bio/files/bowtie2/2.1.0/
Copy bowtie2's executable files (bowtie2, bowtie2-align,	bowtie2-align-debug) 
proper to your linux system to fusionscan/src_dir/
$ cp bowtie2 fusionscan/src_dir/
$ cp bowtie2-align fusionscan/src_dir/
$ cp bowtie2-align-debug	fusionscan/src_dir/

2. SSAHA2 : http://www.sanger.ac.uk/resources/software/ssaha2/
Copy SSAHA2's executalbe	files (ssaha2, ssaha2Build) 
proper to your linux system to fusionscan/src_dir/
$ cp ssaha2 fusionscan/src_dir/
$ cp ssaha2Build	fusionscan/src_dir/

3. FASTX-Toolkit : http://hannonlab.cshl.edu/fastx_toolkit/download.html
Copy FASTX-Toolkit's executable file (fastq_quality_trimmer, fastx_collapser, fastx_artifacts_filter) 
proper to your linux system to fusionscan/src_dir/
$ cp fastq_quality_trimmer fusionscan/src_dir/
$ cp fastx_collapser fusionscan/src_dir/
$ cp fastx_artifacts_filter fusionscan/src_dir/

4. blast (bl2seq) : ftp://ftp.ncbi.nlm.nih.gov/blast/executables/release/LATEST/ 
Copy blast's executable file (bl2seq) 
proper to your linux system to fusionscan/src_dir/
$ cp bl2seq fusionscan/src_dir/

5. blat : http://hgdownload.cse.ucsc.edu/admin/exe/
Copy blat's executable file (gfClient, gfServer,	nibFrag) 
proper to your linux system to fusionscan/src_dir/
$ cp gfClient fusionscan/src_dir/
$ cp gfServer fusionscan/src_dir/
$ cp nibFrag fusionscan/src_dir/

*** cf1.	all index files	for two	alignment program are included in fusionscan/src_dir/)
*** cf2.	in case	of linux_ix86_64 : all binaries	are included in	fusionscan/src_dir/

COMMANDS

Usage:
python Run_FusionScan.py <reads1[,reads2]> <output-dir>	<read-length> <gfserver	host:port> [options]

** <gfserver host:port>	is need	to use blat (gfclient) at the last step.

Options:
--phred33 : qualities are Phred+33 [ default]
--phred64 : qualities are Phred+64
-P/--threads : number of threads <int> [ default: 1]
-ms/--min-seed : minimum number of seed reads <int> [ default: 2]
-md/--min-distance : minimum distance between two genes <int> [ default: 100000 ]
-kfgs/--known-fusion-gene-search : searching known kinase contained fusion genes

Example:
	python	Run_FusionScan.py Test.fa Test 75 00.00.00.00:8100 --phred33 -P 1 -ms 2

OUTPUTS

1. FUSION INFORMATION FILE
	fusionscan/project_name/all_candidates_for_project_name_seed2.txt 
	-------------------------------------------------------
	| column         | explanation             | example  |
	--------------------------------------------------------
	| Hgene          | head gene symbol        | EML4     |
	| Hchr           | head gene's chromosome  | chr2     |
	| Hstrand        | head gene's strnad      | +        |
	| Hbp            | head gene's break point | 42522656 |
	| Tgene          | tail gene symbol        | ALK      |
	| Tchr           | tail gene's chromosome  | chr2     | 
	| Tstrand        | tail gene's strand      | -        |
	| Tbp            | tail gene's break point | 29446394 |
	| Hg-Tg          | head-tail gene pair     | EML4-ALK |
	| # of seed read | seed read count         | 13       |
	| BPseq          | 60bp sequence before & after of exon junction break point | TAGAGCCCACACCTGGGAAAGGACCTAAAGtgtaccgccggaagcaccaggagctgcaag |
	-------------------------------------------------------

2. READ ALIGNMENT FILE FOR EACH FUSION CANDIDATE
	fusionscan/project_name/alignment_dir/EML4_ALK.RNAalign

3. CREATED DIRECTORIES
	- fusionscan/project_name/: is created when you run Run_FusionScan.py.
				 under all directories and files are created in project_name directory.

	- alignment_dir/ (*.RNAalign*): contains the alignment files for fusion candidate reads, 
					 these are created when the total analysis is end.

	- fa_dir/: contains the splited fasta files.

	- pslx_dir/: contains the pslx formated alignment files.

	- log_dir/: contains the log files.

	- all_candidates_for_project_name_seed1.txt: result report for all candidates.

	- all_candidates_for_project_name_seed2.txt: result report for candidates having more than 2 seed reads.

	- cout_per_each_step_project_name: statistics for each step.

	- log_project_name: logs for each reads during running FusionScan.jar


4. COVERAGE PLOT FOR FUSON GENE

	Usage: python make_cvg_plot.py <FusionScan_result_file> <RNA_seq_bamfile>

	** FusionScan's result as same as upper format , especially cases seed >= 2
	** RNA-seq has to be aligned on hg19. This should be input by user.

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