def detect(name, ipfile, bgfile, mapfile, alpha, l_thresh, pickle_file,
wav_file):
"""
This function drives the peak detection workflow.
"""
# set logfile
logfile = name + '.log'
pique.msg(logfile, 'starting run for project : ' + name)
# log inputs
pique.msg(logfile, ' -> IP file : ' + ipfile)
pique.msg(logfile, ' -> BG file : ' + bgfile)
pique.msg(logfile, ' -> map file : ' + mapfile)
pique.msg(logfile, ' -> alpha : ' + str(alpha))
pique.msg(logfile, ' -> l_thresh : ' + str(l_thresh))
# load the data
pique.msg(logfile, 'loading data...')
D = pique.data.PiqueData(ipfile, bgfile, mapfile, name=name)
pique.msg(logfile, ' found contigs :')
for contig in D.data.keys():
pique.msg(logfile, ' ' + contig)
pique.msg(logfile, ' length : ' + str(D.data[contig]['length']))
for r in D.data[contig]['regions']:
start = str(r['start'])
stop = str(r['stop'])
pique.msg(logfile, ' analysis region : ' + start + ':' + stop)
for m in D.data[contig]['masks']:
start = str(m['start'])
stop = str(m['stop'])
pique.msg(logfile, ' masking region : ' + start + ':' + stop)
# start analysis workbench
pique.msg(logfile, 'creating analysis workbench...')
PA = pique.analysis.PiqueAnalysis(D)
# run filters
pique.msg(logfile, 'running filters...')
for ar_name in PA.data.keys():
pique.msg(logfile,
' :: applying filters to analysis region ' + ar_name)
PA.apply_filter(ar_name, alpha, l_thresh)
# find peaks
pique.msg(logfile, 'finding peaks...')
for ar_name in PA.data.keys():
PA.find_peaks(ar_name)
pique.msg(
logfile,
' peaks ' + ar_name + ' : ' + str(len(PA.data[ar_name]['peaks'])))
pique.msg(
logfile,
' noise threshold : ' + str(PA.data[ar_name]['N_thresh']))
pique.msg(
logfile,
' filter threshold : ' + str(PA.data[ar_name]['n_thresh']))
pique.msg(
logfile, ' normalizations : ' +
', '.join(map(str, PA.data[ar_name]['norms'])))
# if a pickle file was requested, write it
if pickle_file:
pique.msg(logfile, 'pickling analysis workbench...')
cPickle.dump(PA, open(name + '.pickle', 'w'))
# if a WAV file was requested, write it
if wav_file:
for contig in D.data.keys():
file = name + '_' + contig + '.wav'
pique.msg(logfile, 'writing WAV output : ' + file)
pique.fileIO.writeWAV(file,
D.data,
contig,
track='IP',
minusBG=True,
amplify=True)
# write output files
pique.msg(logfile, 'writing output files...')
pique.fileIO.writepeaksGFF(name + '.gff', PA.data)
pique.fileIO.writebookmarks(name + '.bookmark', PA.data, name=name)
pique.fileIO.writeQP(name + '.qp', PA.data)
pique.fileIO.writepeakTSV(name + '.peak.tsv', PA.data)
pique.fileIO.writetrack(name + '.IP.track', D.data)
pique.fileIO.writetrack(name + '.BG.track', D.data, track='BG')
# done!
pique.msg(logfile, 'run completed.')
'masking_loci', \
'peak_bookmarks', \
'weed_bookmarks', \
'overlap_track', \
'binding_track' ]
opt_dict = yaml.load( open( sys.argv[1] ).read() )
for opt in num_opts + 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 the 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 )
# apply mask
pique.msg( 'applying mask...' )
is_elements = []
for line in open( masking_loci ) :
if line.__contains__('#') :
continue
start, stop = map( int, line.split()[:2] )
is_elements.append( { 'start':start, 'stop':stop } )
data_ff = pique.mask( data_ff, is_elements )
def makemap(name, bamfile, window, stride, highest, lowest, bins):
"""
This function drives the genome map making workflow.
"""
import pylab
logfile = name + '.mapmaker.log'
mapfile = name + '.map.gff'
pique.msg(logfile, 'starting mapmaker for project : ' + name)
pique.msg(logfile, ' -> BAM file : ' + bamfile)
pique.msg(logfile, ' -> map file : ' + mapfile)
pique.msg(logfile, ' -> window : ' + str(window))
pique.msg(logfile, ' -> stride : ' + str(stride))
pique.msg(logfile, ' -> bins : ' + str(bins))
pique.msg(logfile, ' -> highest bin : ' + str(highest))
pique.msg(logfile, ' -> lowest bin : ' + str(lowest))
pique.msg(logfile, 'loading data...')
data = pique.fileIO.loadBAM(bamfile)
pique.msg(logfile, ' found contigs :')
for contig in data.keys():
pique.msg(logfile, ' ' + contig)
pique.msg(logfile, ' ' + str(len(data[contig]['forward'])))
pique.msg(logfile, 'making spectral histograms...')
sh = {}
for contig in data.keys():
pique.msg(logfile,
' :: making sectral histogram for contig ' + contig)
d = numpy.array(data[contig]['forward'] + data[contig]['reverse'],
dtype=int)
sh[contig] = pique.mapmaker.hist(d, lowest, highest, bins, window,
stride)
# save images of spectral histograms
pique.msg(logfile, 'saving images of spectral histograms...')
for contig in sh.keys():
pylab.cla() # clean up crumbs from last plot
pylab.clf() # clean up crumbs from last plot
pique.msg(logfile, ' :: saving image for contig ' + contig)
pylab.contourf(sh[contig], bins)
pylab.title(name + ' : ' + contig)
imgname = name + '_' + contig + '.png'
pylab.savefig(imgname, format='png')
def bam2wav(name, ipfile, bgfile):
"""
This function drives the creation of a WAV file from a BAM file.
"""
# set logfile
logfile = name + '.log'
pique.msg(logfile,
'converting BAM files to WAV files for project : ' + name)
# log inputs
pique.msg(logfile, ' -> IP file : ' + ipfile)
pique.msg(logfile, ' -> BG file : ' + bgfile)
# load the data
pique.msg(logfile, 'loading data...')
D = pique.data.PiqueData(ipfile, bgfile, '', name=name)
pique.msg(logfile, ' found contigs :')
for contig in D.data.keys():
pique.msg(logfile, ' ' + contig)
pique.msg(logfile, ' length : ' + str(D.data[contig]['length']))
for r in D.data[contig]['regions']:
start = str(r['start'])
stop = str(r['stop'])
pique.msg(logfile, ' analysis region : ' + start + ':' + stop)
for m in D.data[contig]['masks']:
start = str(m['start'])
stop = str(m['stop'])
pique.msg(logfile, ' masking region : ' + start + ':' + stop)
# write the WAV files
for contig in D.data.keys():
file = name + '_' + contig + '.wav'
pique.msg(logfile, 'writing WAV output : ' + file)
pique.fileIO.writeWAV(file,
D.data,
contig,
track='IP',
minusBG=True,
amplify=True)
# done!
pique.msg(logfile, 'conversion completed.')
def makemap( name,
bamfile,
window,
stride,
highest,
lowest,
bins ) :
"""
This function drives the genome map making workflow.
"""
import pylab
logfile = name + '.mapmaker.log'
mapfile = name + '.map.gff'
pique.msg( logfile, 'starting mapmaker for project : ' + name )
pique.msg( logfile, ' -> BAM file : ' + bamfile )
pique.msg( logfile, ' -> map file : ' + mapfile )
pique.msg( logfile, ' -> window : ' + str(window) )
pique.msg( logfile, ' -> stride : ' + str(stride) )
pique.msg( logfile, ' -> bins : ' + str(bins) )
pique.msg( logfile, ' -> highest bin : ' + str(highest) )
pique.msg( logfile, ' -> lowest bin : ' + str(lowest) )
pique.msg( logfile, 'loading data...' )
data = pique.fileIO.loadBAM( bamfile )
pique.msg( logfile, ' found contigs :' )
for contig in data.keys() :
pique.msg( logfile, ' ' + contig )
pique.msg( logfile, ' ' + str(len(data[contig]['forward'])) )
pique.msg( logfile, 'making spectral histograms...' )
sh = {}
for contig in data.keys() :
pique.msg( logfile, ' :: making sectral histogram for contig ' + contig )
d = numpy.array( data[contig]['forward'] + data[contig]['reverse'], dtype = int )
sh[contig] = pique.mapmaker.hist( d,
lowest,
highest,
bins,
window,
stride )
# save images of spectral histograms
pique.msg( logfile, 'saving images of spectral histograms...' )
for contig in sh.keys() :
pylab.cla() # clean up crumbs from last plot
pylab.clf() # clean up crumbs from last plot
pique.msg( logfile, ' :: saving image for contig ' + contig )
pylab.contourf( sh[contig], bins )
pylab.title( name + ' : ' + contig )
imgname = name + '_' + contig + '.png'
pylab.savefig( imgname, format='png' )
def filter_all( self, alpha, l_thresh ) :
for ar_name in self.data.keys() :
pique.msg( ' :: applying filters to analysis region ' + ar_name )
self.apply_filter( ar_name, alpha, l_thresh )
'forward_bgnd_track', \
'reverse_ChIP_track', \
'reverse_bgnd_track', \
'slice_bookmarks', \
'new_track_prefix', ]
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 the 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
pique.msg( 'reading annotations...' )
slices = pique.readbookmarks( slice_bookmarks )
# write new slice tracks
for s in slices :
sdata_ff = data_ff[ s['start'] : s['stop'] ]
sdata_rr = data_rr[ s['start'] : s['stop'] ]
sb_ff = b_ff[ s['start'] : s['stop'] ]
sb_rr = b_rr[ s['start'] : s['stop'] ]
def bam2wav( name,
ipfile,
bgfile ) :
"""
This function drives the creation of a WAV file from a BAM file.
"""
# set logfile
logfile = name + '.log'
pique.msg( logfile, 'converting BAM files to WAV files for project : ' + name )
# log inputs
pique.msg( logfile, ' -> IP file : ' + ipfile )
pique.msg( logfile, ' -> BG file : ' + bgfile )
# load the data
pique.msg( logfile, 'loading data...' )
D = pique.data.PiqueData( ipfile, bgfile, '', name=name )
pique.msg( logfile, ' found contigs :' )
for contig in D.data.keys() :
pique.msg( logfile, ' ' + contig )
pique.msg( logfile, ' length : ' + str(D.data[contig]['length']) )
for r in D.data[contig]['regions'] :
start = str( r['start'] )
stop = str( r['stop'] )
pique.msg( logfile, ' analysis region : ' + start + ':' + stop )
for m in D.data[contig]['masks'] :
start = str( m['start'] )
stop = str( m['stop'] )
pique.msg( logfile, ' masking region : ' + start + ':' + stop )
# write the WAV files
for contig in D.data.keys() :
file = name + '_' + contig + '.wav'
pique.msg( logfile, 'writing WAV output : ' + file )
pique.fileIO.writeWAV( file,
D.data,
contig,
track='IP',
minusBG=True,
amplify=True )
# done!
pique.msg( logfile, 'conversion completed.' )
'forward_bgnd_track', \
'reverse_ChIP_track', \
'reverse_bgnd_track', \
'slice_bookmarks', \
'new_track_prefix', ]
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 the 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
pique.msg('reading annotations...')
slices = pique.readbookmarks(slice_bookmarks)
# write new slice tracks
for s in slices:
sdata_ff = data_ff[s['start']:s['stop']]
sdata_rr = data_rr[s['start']:s['stop']]
sb_ff = b_ff[s['start']:s['stop']]
sb_rr = b_rr[s['start']:s['stop']]
'forward_bgnd_track', \
'reverse_ChIP_track', \
'reverse_bgnd_track', \
'masking_loci', \
'annotated_bookmarks' ]
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)
'forward_bgnd_track', \
'reverse_ChIP_track', \
'reverse_bgnd_track', \
'new_forward_bgnd_track', \
'new_reverse_bgnd_track', \
'non_peak_bookmarks', ]
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] )
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 )
non_peaks = pique.readbookmarks( non_peak_bookmarks )
# calculate enrichment ratios
pique.msg( 'calculating enrichment ratios using ' + \
str(len(non_peaks)) + 'regions...' )
d_f, d_r, b_f, b_r = [],[],[],[]
for n,region in enumerate( non_peaks ) :
d_f.append( sum( data_ff[ region['start'] : region['stop'] ] ) )
d_r.append( sum( data_rr[ region['start'] : region['stop'] ] ) )
def run( self ) :
# check inputs...
name = self.nametext.get().strip()
# set logfile
logfile = name + '.log'
pique.msg( logfile, 'starting run for project : ' + name )
alpha = int( self.alphatext.get().strip() )
l_thresh = int( self.lthreshtext.get().strip() )
# log inputs
pique.msg( logfile, ' -> IP file : ' + self.IPfile )
pique.msg( logfile, ' -> BG file : ' + self.BGfile )
pique.msg( logfile, ' -> map file : ' + self.mapfile )
pique.msg( logfile, ' -> alpha : ' + str(alpha) )
pique.msg( logfile, ' -> l_thresh : ' + str(l_thresh) )
# load the data
pique.msg( logfile, 'loading data...' )
self.master.title( 'Pique : loading data...' )
if not self.mapfile :
D = pique.data.PiqueData( self.IPfile, self.BGfile, name=name )
else :
D = pique.data.PiqueData( self.IPfile, self.BGfile, self.mapfile, name=name )
pique.msg( logfile, ' found contigs :' )
for contig in D.data.keys() :
pique.msg( logfile, ' ' + contig )
pique.msg( logfile, ' length : ' + str(D.data[contig]['length']) )
for r in D.data[contig]['regions'] :
start = str( r['start'] )
stop = str( r['stop'] )
pique.msg( logfile, ' analysis region : ' + start + ':' + stop )
for m in D.data[contig]['masks'] :
start = str( m['start'] )
stop = str( m['stop'] )
pique.msg( logfile, ' masking region : ' + start + ':' + stop )
# start analysis workbench
pique.msg( logfile, 'creating analysis workbench...' )
self.master.title( 'Pique : creating workbench...' )
PA = pique.analysis.PiqueAnalysis( D )
# run filters
pique.msg( logfile, 'running filters...' )
self.master.title( 'Pique : running filters...' )
for ar_name in PA.data.keys() :
pique.msg( logfile, ' :: applying filters to analysis region ' + ar_name )
PA.apply_filter( ar_name, alpha, l_thresh )
# find peaks
pique.msg( logfile, 'finding peaks...' )
self.master.title( 'Pique : finding peaks...' )
for ar_name in PA.data.keys() :
PA.find_peaks(ar_name)
pique.msg( logfile, ' peaks ' + ar_name + ' : ' + str(len(PA.data[ar_name]['peaks'])) )
pique.msg( logfile, ' noise threshold : ' + str(PA.data[ar_name]['N_thresh']) )
pique.msg( logfile, ' filter threshold : ' + str(PA.data[ar_name]['n_thresh']) )
pique.msg( logfile, ' normalizations : ' + ', '.join( map(str, PA.data[ar_name]['norms']) ) )
# write output files
pique.msg( logfile, 'writing output files...' )
self.master.title( 'Pique : writing output...' )
pique.fileIO.writepeaksGFF( name + '.gff', PA.data )
pique.fileIO.writebookmarks( name + '.bookmark', PA.data, name=name )
pique.fileIO.writeQP( name + '.qp', PA.data )
pique.fileIO.writepeakTSV( name + '.peak.tsv', PA.data )
pique.fileIO.writetrack( name + '.IP.track', D.data )
pique.fileIO.writetrack( name + '.BG.track', D.data, track='BG' )
# done!
pique.msg( logfile, 'run completed.' )
self.master.title( 'Pique : run completed.' )
def run( self ) :
# check inputs...
name = self.nametext.get().strip()
pique.msg( 'starting run for project : ' + name )
alpha = int( self.alphatext.get().strip() )
l_thresh = int( self.lthreshtext.get().strip() )
# load the data
pique.msg( 'loading data...' )
self.master.title( 'Pique : loading data...' )
if not self.mapfile :
D = pique.data.PiqueData( self.IPfile, self.BGfile )
else :
D = pique.data.PiqueData( self.IPfile, self.BGfile, self.mapfile )
pique.msg( ' -> found contigs :' )
for contig in D.data.keys() :
pique.msg( ' ' + contig )
pique.msg( ' length : ' + str(D.data[contig]['length']) )
for r in D.data[contig]['regions'] :
start = str( r['start'] )
stop = str( r['stop'] )
pique.msg( ' analysis region : ' + start + ':' + stop )
for m in D.data[contig]['masks'] :
start = str( m['start'] )
stop = str( m['stop'] )
pique.msg( ' masking region : ' + start + ':' + stop )
# start analysis workbench
pique.msg( 'creating analysis workbench...' )
self.master.title( 'Pique : creating workbench...' )
PA = pique.analysis.PiqueAnalysis( D )
# run filters
pique.msg( 'running filters...' )
self.master.title( 'Pique : running filters...' )
pique.msg( ' -> alpha : ' + str(alpha) )
pique.msg( ' -> l_thresh : ' + str(l_thresh) )
PA.filter_all( alpha, l_thresh )
# find peaks
pique.msg( 'finding peaks...' )
self.master.title( 'Pique : finding peaks...' )
for ar_name in PA.data.keys() :
PA.find_peaks(ar_name)
pique.msg( ' :: ' + ar_name + ' : ' + str(len(PA.data[ar_name]['peaks'])) )
# write output files
pique.msg( 'writing output files...' )
self.master.title( 'Pique : writing output...' )
pique.fileIO.writepeaksGFF( name + '.gff', PA.data )
pique.fileIO.writebookmarks( name + '.bookmark', PA.data )
# done!
pique.msg( 'run completed.' )
self.master.title( 'Pique : run completed.' )
'forward_bgnd_track', \
'reverse_ChIP_track', \
'reverse_bgnd_track', \
'new_forward_bgnd_track', \
'new_reverse_bgnd_track', \
'non_peak_bookmarks', ]
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])
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)
non_peaks = pique.readbookmarks(non_peak_bookmarks)
# calculate enrichment ratios
pique.msg( 'calculating enrichment ratios using ' + \
str(len(non_peaks)) + 'regions...' )
d_f, d_r, b_f, b_r = [], [], [], []
for n, region in enumerate(non_peaks):
d_f.append(sum(data_ff[region['start']:region['stop']]))
d_r.append(sum(data_rr[region['start']:region['stop']]))
'masking_loci', \
'peak_bookmarks', \
'weed_bookmarks', \
'overlap_track', \
'binding_track' ]
opt_dict = yaml.load(open(sys.argv[1]).read())
for opt in num_opts + 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 the 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)
# apply mask
pique.msg('applying mask...')
is_elements = []
for line in open(masking_loci):
if line.__contains__('#'):
continue
start, stop = map(int, line.split()[:2])
is_elements.append({'start': start, 'stop': stop})
data_ff = pique.mask(data_ff, is_elements)
'forward_bgnd_track', \
'reverse_ChIP_track', \
'reverse_bgnd_track', \
'masking_loci', \
'annotated_bookmarks' ]
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)
opt_dict = yaml.load( open( sys.argv[1] ).read() )
for opt in num_opts + str_opts :
if not opt_dict.has_key( opt ) :
print 'config file missing option : ' + opt
quit()
setattr( sys.modules[__name__], opt, opt_dict[opt] )
for opt in opt_opts :
if opt_dict.has_key( opt ) :
setattr( sys.modules[__name__], opt, opt_dict[opt] )
else :
setattr( sys.modules[__name__], opt, None )
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 )
pique.msg( 'applying mask...' )
is_elements = []
for line in open( masking_loci ) :
if line.__contains__('#') :
continue
start, stop = map( int, line.split()[:2] )
is_elements.append( { 'start':start, 'stop':stop } )
data_ff = pique.mask( data_ff, is_elements )
data_rr = pique.mask( data_rr, is_elements )
def detect( name,
ipfile,
bgfile,
mapfile,
alpha,
l_thresh,
pickle_file,
wav_file ) :
"""
This function drives the peak detection workflow.
"""
# set logfile
logfile = name + '.log'
pique.msg( logfile, 'starting run for project : ' + name )
# log inputs
pique.msg( logfile, ' -> IP file : ' + ipfile )
pique.msg( logfile, ' -> BG file : ' + bgfile )
pique.msg( logfile, ' -> map file : ' + mapfile )
pique.msg( logfile, ' -> alpha : ' + str(alpha) )
pique.msg( logfile, ' -> l_thresh : ' + str(l_thresh) )
# load the data
pique.msg( logfile, 'loading data...' )
D = pique.data.PiqueData( ipfile, bgfile, mapfile, name=name )
pique.msg( logfile, ' found contigs :' )
for contig in D.data.keys() :
pique.msg( logfile, ' ' + contig )
pique.msg( logfile, ' length : ' + str(D.data[contig]['length']) )
for r in D.data[contig]['regions'] :
start = str( r['start'] )
stop = str( r['stop'] )
pique.msg( logfile, ' analysis region : ' + start + ':' + stop )
for m in D.data[contig]['masks'] :
start = str( m['start'] )
stop = str( m['stop'] )
pique.msg( logfile, ' masking region : ' + start + ':' + stop )
# start analysis workbench
pique.msg( logfile, 'creating analysis workbench...' )
PA = pique.analysis.PiqueAnalysis( D )
# run filters
pique.msg( logfile, 'running filters...' )
for ar_name in PA.data.keys() :
pique.msg( logfile, ' :: applying filters to analysis region ' + ar_name )
PA.apply_filter( ar_name, alpha, l_thresh )
# find peaks
pique.msg( logfile, 'finding peaks...' )
for ar_name in PA.data.keys() :
PA.find_peaks(ar_name)
pique.msg( logfile, ' peaks ' + ar_name + ' : ' + str(len(PA.data[ar_name]['peaks'])) )
pique.msg( logfile, ' noise threshold : ' + str(PA.data[ar_name]['N_thresh']) )
pique.msg( logfile, ' filter threshold : ' + str(PA.data[ar_name]['n_thresh']) )
pique.msg( logfile, ' normalizations : ' + ', '.join( map(str, PA.data[ar_name]['norms']) ) )
# if a pickle file was requested, write it
if pickle_file :
pique.msg( logfile, 'pickling analysis workbench...' )
cPickle.dump( PA, open( name + '.pickle', 'w' ) )
# if a WAV file was requested, write it
if wav_file :
for contig in D.data.keys() :
file = name + '_' + contig + '.wav'
pique.msg( logfile, 'writing WAV output : ' + file )
pique.fileIO.writeWAV( file,
D.data,
contig,
track='IP',
minusBG=True,
amplify=True )
# write output files
pique.msg( logfile, 'writing output files...' )
pique.fileIO.writepeaksGFF( name + '.gff', PA.data )
pique.fileIO.writebookmarks( name + '.bookmark', PA.data, name=name )
pique.fileIO.writeQP( name + '.qp', PA.data )
pique.fileIO.writepeakTSV( name + '.peak.tsv', PA.data )
pique.fileIO.writetrack( name + '.IP.track', D.data )
pique.fileIO.writetrack( name + '.BG.track', D.data, track='BG' )
# done!
pique.msg( logfile, 'run completed.' )
#!/usr/bin/env python
"""
A very, very basic genome browser.
usage : ./basic_browser.py forward.track \
reverse.track \
peak.bookmark \
genes.bed
"""
import numpy
import pique
import pylab
import sys
# read track data
pique.msg('reading track data...')
data_ff = pique.readtrack(sys.argv[1])
data_rr = pique.readtrack(sys.argv[2])
# read peak data
pique.msg('reading peak bookmarks...')
peaks = pique.readbookmarks(sys.argv[3])
# draw peaks
pique.msg('drawing peak bookmarks...')
for peak in peaks:
axvspan(peak['start'], peak['stop'], color='green', alpha=0.3)
# read BED formatted gene annotations
pique.msg('reading gene annotations...')
genes = {}
#!/usr/bin/env python
"""
A very, very basic genome browser.
usage : ./basic_browser.py forward.track \
reverse.track \
peak.bookmark \
genes.bed
"""
import numpy
import pique
import pylab
import sys
# read track data
pique.msg( 'reading track data...' )
data_ff = pique.readtrack( sys.argv[1] )
data_rr = pique.readtrack( sys.argv[2] )
# read peak data
pique.msg( 'reading peak bookmarks...' )
peaks = pique.readbookmarks( sys.argv[3] )
# draw peaks
pique.msg( 'drawing peak bookmarks...' )
for peak in peaks :
axvspan( peak['start'], peak['stop'], color='green', alpha=0.3 )
# read BED formatted gene annotations
pique.msg( 'reading gene annotations...' )
genes = {}
str_opts = [ 'track_name', \
'annotated_bookmarks', \
'gene_annotations', \
'new_bookmarks', ]
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 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())
