Instructions for using "peakdefiner.py". A meta-approach for identifying ChIP-Seq peaks 
for transcription factors.

DOWNLOADS

The script is written in Python, and uses Numpy. It is run in linux-environment.
The script is based on output from two other peakfinder programs: 
MACS(Zhang et al, 2008, Genome Biology) v 1.3.7.1. http://liulab.dfci.harvard.edu/MACS/ 
( A new version of MACS is available. The script has not been tested on the new version, however it should run if the 
outputfiles contain the first three columns 'chrom - startpos - endpos') 
SISSRs(Jothi et al, 2008, Nucleic Acids Research), v 1.4. http://sissrs.rajajothi.com/

MACS and SISSRs must be installed for the script to run.

COMMAND LINE

python peak_definer.py -s=sample.bed -b=background.bed -r=replicate.bed -bal=yes -mf=32

INPUT PARAMETERS

      -s:			filename of ChIP-Seq sample. Must be in Bed-format. 
      -b: (optional)		filename of background sample. Must be in Bed-format.
      -r: (optional)		filename of replicate sample. Must be in Bed-format.
      -bal: (optional)(yes/no)	Whether to balance the total sample and replicate tags towards the background tags. Default is "yes"
      -mf: (optional)		The "mfold" parameter from MACS. Default (from MACS) is 32. Lower this value if MACS do not find 
      	   			enough peaks to build the shift-model.

OUTPUT FILES

       Main output:       

       peaks_file.bed:		Final trimmed peaks
       regions_file.bed:	Final regions before trimming

       Columns in peaks_file.bed are: chromosome, startpos, endpos, peak-length, number of tags (including non-unique tags).

       Other generated files:
       
       bal_bkg.bed:		Balanced set of background tags based on random sampling
       macs_peaks.xls:		Sample peaks from MACS
       sisr_peaks.txt:		Sample peaks from SISSRs
       macs_peaks_rep.xls:	Replicate peaks from MACS
       "sample".pdf		sample pdf shift-model from MACS
       "replicate".pdf		replicate pdf shift-model from MACS