opt_dict = yaml.load(open(sys.argv[1]).read())
for opt in str_opts:
if not opt_dict.has_key(opt):
print 'config file missing option : ' + opt
quit()
setattr(sys.modules[__name__], opt, opt_dict[opt])
# read track data
pique.msg('reading track data...')
data_ff = pique.readtrack(forward_ChIP_track)
data_rr = pique.readtrack(reverse_ChIP_track)
b_ff = pique.readtrack(forward_bgnd_track)
b_rr = pique.readtrack(reverse_bgnd_track)
# read bookmarks file
peaks = pique.readbookmarks(peak_bookmarks)
# calculate enrichment ratios
for n, peak in enumerate(peaks):
a = sum(data_ff[peak['start']:peak['stop']])
a = a + sum(data_rr[peak['start']:peak['stop']])
b = sum(b_ff[peak['start']:peak['stop']])
b = b + sum(b_rr[peak['start']:peak['stop']])
peaks[n]['annotations']['enrichment_ratio'] = float(a) / float(b)
# write new bookmark file
pique.msg('writing re-annotated bookmark file...')
pique.writebookmarks(annotated_bookmarks, track_name, peaks)
for n,peak in enumerate( envelope ) :
df = data_ff[ peak['forward']['start'] : peak['reverse']['stop'] ]
dr = data_rr[ peak['forward']['start'] : peak['reverse']['stop'] ]
df = scipy.signal.convolve( df, numpy.ones(10) / 10.0 )
dr = scipy.signal.convolve( dr, numpy.ones(10) / 10.0 )
c = peak['forward']['start'] + ( dr.argmax() + df.argmax() ) / 2.0
envelope[n]['binds'] = int(c)
for n in range(len(weeds)) :
weeds[n]['binds'] = ''
# write output files
pique.msg( 'writing output files...' )
peaks_f = numpy.zeros( len(data_f) )
peaks_r = numpy.zeros( len(data_r) )
for peak in forward :
peaks_f[ peak['start'] : peak['stop'] ] = 1
for peak in reverse :
peaks_r[ peak['start'] : peak['stop'] ] = 1
pique.write_strandless_track( peaks_f+peaks_r, overlap_track, track_name )
pique.writebookmarks( peak_bookmarks, track_name, envelope )
pique.writebookmarks( weed_bookmarks, track_name, weeds )
f = open( binding_track, 'w' )
f.write( 'sequence\tstrand\tposition\tvalue\n' )
for peak in envelope :
f.write( track_name + '\t.\t' + str(peak['binds']) + '\t1\n' )
f.close()
opt_dict = yaml.load( open( sys.argv[1] ).read() )
for opt in str_opts :
if not opt_dict.has_key( opt ) :
print 'config file missing option : ' + opt
quit()
setattr( sys.modules[__name__], opt, opt_dict[opt] )
# read track data
pique.msg( 'reading track data...' )
data_ff = pique.readtrack( forward_ChIP_track )
data_rr = pique.readtrack( reverse_ChIP_track )
b_ff = pique.readtrack( forward_bgnd_track )
b_rr = pique.readtrack( reverse_bgnd_track )
# read bookmarks file
peaks = pique.readbookmarks( peak_bookmarks )
# calculate enrichment ratios
for n,peak in enumerate(peaks) :
a = sum( data_ff[ peak['start'] : peak['stop'] ] )
a = a + sum( data_rr[ peak['start'] : peak['stop'] ] )
b = sum( b_ff[ peak['start'] : peak['stop'] ] )
b = b + sum( b_rr[ peak['start'] : peak['stop'] ] )
peaks[n]['annotations']['enrichment_ratio'] = float(a) / float(b)
# write new bookmark file
pique.msg( 'writing re-annotated bookmark file...' )
pique.writebookmarks( annotated_bookmarks, track_name, peaks )
for n, peak in enumerate(envelope):
df = data_ff[peak['forward']['start']:peak['reverse']['stop']]
dr = data_rr[peak['forward']['start']:peak['reverse']['stop']]
df = scipy.signal.convolve(df, numpy.ones(10) / 10.0)
dr = scipy.signal.convolve(dr, numpy.ones(10) / 10.0)
c = peak['forward']['start'] + (dr.argmax() + df.argmax()) / 2.0
envelope[n]['binds'] = int(c)
for n in range(len(weeds)):
weeds[n]['binds'] = ''
# write output files
pique.msg('writing output files...')
peaks_f = numpy.zeros(len(data_f))
peaks_r = numpy.zeros(len(data_r))
for peak in forward:
peaks_f[peak['start']:peak['stop']] = 1
for peak in reverse:
peaks_r[peak['start']:peak['stop']] = 1
pique.write_strandless_track(peaks_f + peaks_r, overlap_track, track_name)
pique.writebookmarks(peak_bookmarks, track_name, envelope)
pique.writebookmarks(weed_bookmarks, track_name, weeds)
f = open(binding_track, 'w')
f.write('sequence\tstrand\tposition\tvalue\n')
for peak in envelope:
f.write(track_name + '\t.\t' + str(peak['binds']) + '\t1\n')
f.close()
# read bookmarks file
pique.msg( 'reading annotations...' )
peaks = pique.readbookmarks( annotated_bookmarks )
# read gene annotations
genes = {}
for line in open( gene_annotations ) :
if line.__contains__( '\"' ) :
continue
if not line.split()[0].lower() == track_name :
continue
start,stop = map( int, line.split()[1:3] )
strand = line.strip().split()[5]
name = line.split()[3]
genes[name] = { 'start':start,'stop':stop,'strand':strand }
print name
print len(genes.keys())
# add gene annotation to peak list
pique.msg( 'finding transcript start sites in enriched regions...' )
for peak in peaks :
gg = []
for name,gene in genes.items() :
if peak['start'] < gene['start'] < peak['stop'] :
gg.append(name)
peak['annotations']['genes'] = ','.join(gg)
# write new bookmark file
pique.msg( 'writing re-annotated bookmark file...' )
pique.writebookmarks( new_bookmarks, track_name, peaks )