FineSplice Files

    
    FineSplice, a Python wrapper to TopHat2 for enhanced splice 
    junction detection and quantification from RNA-Seq data


USAGE
        
    python FineSplice.py -i <filename>.bam -l <read length>

        -i  <filename>.bam	  (path to TopHat2 BAM file)
        -l  <read length> 	  (read length)
    
    e.g. python FineSplice.py -i example.bam -l 50


PREREQUISITES
    
TopHat2 alignment in BAM format

    Align with reference transcript annotations (i.e. -G/--GTF option)    

Python 2.x (version >= 2.6) with the following modules installed:

	pysam          (version tested 0.7.4)
	scikit-learn   (version tested 0.13.1)
	numpy          (version tested 1.7.1, required by scikit-learn)
	scipy          (version tested 0.7.2, required by scikit-learn)
	
Get the latest versions at:

    http://code.google.com/p/pysam/     (pysam)
    http://scikit-learn.org/            (scikit-learn)


OUTPUT FORMAT

FineSplice produces two output files:

    <filename>.accepted.junc
    <filename>.discarded.junc
    
Consisting of a tab-separated list of splice junctions with the following fields:
    
  1.  SN         (reference sequence name i.e. chromosome)
  2.  start      (junction start genomic coordinate,0-based)
  3.  end        (junction end genomic coordinate, 0-based) 
  4.  prob       (posterior probability i.e. junction reliability)
  5.  unique     (# uniquely mapping reads)            
[ 6.  rescued    (# multiple mapping reads rescued after filtering) ]     
 

DESCRIPTION        
    
    FineSplice is a post-processing tool geared towards a
    a reliable identification of expressed exon junctions
    from TopHat2, at enhanced detection precision with
    small loss in sensitivity.
    
    Following alignment with TopHat2 (using known transcript 
    annotations) FineSplice takes as input the resulting BAM 
    file and outputs a confident set of expressed junctions,
    with the corresponding read counts. 
    
    Potential false positives arising from spurious alignments 
    are filtered out through a semi-supervised anomaly detection 
    strategy based on logistic regression. Multiple mapping reads 
    with a unique hit after filtering are rescued and reallocated 
    to the most reliable candidate location. 


REFERENCE

For further details check out our paper:

    Gatto,A. et al (2014) FineSplice, enhanced splice junction 
    detection and quantification: a novel pipeline based on the 
    assessment of diverse RNA-Seq alignment solutions. 

If you use FineSplice in scientific publications, please cite our paper.

Get FineSplice latest version at: https://sourceforge.net/projects/finesplice/