to_sort += ['algorithm', 'peaks']
sort_order = ' '.join([i + '=+' for i in to_sort])
composite = CompositeTrack(
name=hub_config['hub']['name'] + 'composite',
short_label='ChIP-seq composite',
long_label='ChIP-seq composite',
dimensions=dimensions_from_subgroups(subgroups),
filterComposite=filter_composite_from_subgroups(subgroups),
sortOrder=sort_order,
tracktype='bigWig')
signal_view = ViewTrack(name='signalviewtrack',
view='signal',
visibility='full',
tracktype='bigWig',
short_label='signal',
long_label='signal')
peaks_view = ViewTrack(name='peaksviewtrack',
view='peaks',
visibility='dense',
tracktype='bigBed',
short_label='peaks',
long_label='peaks')
supplemental_view = ViewTrack(name='suppviewtrack',
view='supplemental',
visibility='full',
tracktype='bigBed',
short_label='Supplemental',
to_sort += [sg.name for sg in subgroups if sg.name not in to_sort]
sort_order = ' '.join([i + '=+' for i in to_sort])
composite = CompositeTrack(
name=hub_config['hub']['name'] + 'composite',
short_label='rnaseq composite',
long_label='rnaseq composite',
dimensions=dimensions_from_subgroups(subgroups),
filterComposite=filter_composite_from_subgroups(subgroups),
sortOrder=sort_order,
tracktype='bigWig')
# ASSUMPTION: stranded bigwigs
pos_signal_view = ViewTrack(name='possignalviewtrack',
view='possignal',
visibility='full',
tracktype='bigWig',
short_label='pos strand',
long_label='positive strand signal')
neg_signal_view = ViewTrack(name='negsignalviewtrack',
view='negsignal',
visibility='full',
tracktype='bigWig',
short_label='neg strand',
long_label='negative strand signal')
supplemental_view = ViewTrack(name='suppviewtrack',
view='supplementalview',
visibility='full',
tracktype='bigBed',
short_label='Supplemental',
long_label='Supplemental')
sample_dir = config['sample_dir']
samples = df[df.columns[0]]
composite = CompositeTrack(
name=hub_config['hub']['name'] + 'composite',
short_label='rnaseq composite',
long_label='rnaseq composite',
dimensions=dimensions_from_subgroups(subgroups),
filterComposite=filter_composite_from_subgroups(subgroups),
sortOrder=sort_order,
tracktype='bigWig')
# ASSUMPTION: stranded bigwigs
sense_signal_view = ViewTrack(name='sensesignalviewtrack',
view='sensesignal',
visibility='full',
tracktype='bigWig',
short_label='sense strand',
long_label='sense strand signal')
antisense_signal_view = ViewTrack(name='antisensesignalviewtrack',
view='antisensesignal',
visibility='full',
tracktype='bigWig',
short_label='antisense strand',
long_label='antisense strand signal')
supplemental_view = ViewTrack(name='suppviewtrack',
view='supplementalview',
visibility='full',
tracktype='bigBed',
short_label='Supplemental',
long_label='Supplemental')
def main():
parser = argparse.ArgumentParser(
description='Make trackhubs for UCSC browser using bigBed files. \
Outputs to CURRENT DIRECTORY.')
parser.add_argument('inputdir',
metavar='INDIR',
help='Directory containing bigBed files .bb ending')
parser.add_argument('outdir',
metavar='OUTDIR',
help='Directory for staging files')
parser.add_argument('--quiet',
'-q',
action='store_true',
help='Suppress some print statements')
parser.add_argument('--render',
'-r',
action='store_true',
help='Render file to current dir')
parser.add_argument('--upload',
'-u',
action='store_true',
help='Upload file to webserver')
parser.add_argument('--mm9',
'-m',
action='store_true',
help='Switch from mm10 to mm9')
parser.add_argument('--has_strand',
'-s',
action='store_true',
help='Bed has strand (changes from 5 columns to 6)')
parser.add_argument('--suffix',
'-S',
metavar="trackhub label suffix",
default="",
help='Suffix to label, for example H3K4me1')
args = parser.parse_args()
# store command line arguments for reproducibility
CMD_INPUTS = ' '.join(['python'] + sys.argv) # easy printing later
# store argparse inputs for reproducibility / debugging purposes
args_dic = vars(args)
ARG_INPUTS = ['%s=%s' % (key, val) for key, val in args_dic.iteritems()]
# ARG_INPUTS = ['%s=%s' % (key, val) for key, val in args_dic.items()]
ARG_INPUTS = ' '.join(ARG_INPUTS)
# Print arguments supplied by user
if not args.quiet:
print('Command line inputs:')
print(CMD_INPUTS)
print('Argparse variables:')
print(ARG_INPUTS)
# define constants (hard coded)
if args.mm9:
genobuild = "mm9"
else:
genobuild = "mm10"
jsuffix = "%s_%s" % (genobuild, args.suffix)
print("Assigning prefix: %s" % jsuffix)
# dirname: motevo_from_peaks/H3K4me1_peaks
dirname = "motevo_from_peaks/%s_peaks/motevo_motifs_%s" % (args.suffix,
jsuffix)
hubname = "motevo_motifs_%s" % jsuffix
shortlab = "motevo_%s" % jsuffix
longlab = "Motevo motifs %s" % jsuffix
email = "*****@*****.**"
# url = "http://upnaepc2.epfl.ch"
url = "http://upnaesrv1.epfl.ch"
assay = "bigbed"
jvis = "dense"
# bigbed options loaded into ViewTrack
jspectrum = "on"
scoremax = 1000
scoremin = 500
# define URLs
url_main = "%s/%s" % (url, dirname)
url_base = "%s/%s/data" % (url, dirname)
# upload_main = "~/Sites/%s" % dirname
# upload_base = "~/Sites/%s/data" % dirname
upload_main = "%s" % hubname
upload_base = "%s/data" % hubname
if not args.has_strand:
ftype = "bigBed 5"
else:
ftype = "bigBed 6"
# host = "circadian.epfl.ch"
# user = "******"
host = "upnaesrv1.epfl.ch"
user = "******"
# define constants
genomebuild = genobuild
files_dic = get_files_from_dir(args.inputdir, ext=".bb")
samples_dic = {}
for sample in files_dic.keys():
samples_dic[sample] = sample
# init hub genomes file genome trackdb
# Make my hub
hub = Hub(hub=hubname,
short_label=shortlab,
long_label=longlab,
email=email)
# url = "%s/%s" % (url, dirname))
hub.url = os.path.join(url_main, "%s.hub.txt" % hub.hub)
genomes_file = GenomesFile()
genome = Genome(genomebuild)
trackdb = TrackDb()
# add remote fn
# hub.remote_fn = os.path.join(upload_main, "hub.txt")
# genomes_file.remote_fn = os.path.join(upload_main, "genomes.txt")
hub.remote_fn = upload_main
genomes_file.remote_fn = upload_main
trackdb.remote_fn = os.path.join(upload_main, genomebuild, "trackDb.txt")
hub.add_genomes_file(genomes_file)
genome.add_trackdb(trackdb)
genomes_file.add_genome(genome)
# init composite
composite = CompositeTrack(name=hubname,
short_label=shortlab,
long_label=longlab,
tracktype=ftype)
# make subgroups
subgroups = [
SubGroupDefinition(name="sample", label="sample", mapping=samples_dic),
]
composite.add_subgroups(subgroups)
# make viewTrack, a hierarchy containing my files, for example
view = ViewTrack(
name="%sViewTrack" % assay,
view="%s" % assay,
visibility=jvis,
tracktype=ftype,
short_label="%s" % assay,
long_label="%s assay" % assay,
# big bed labels
spectrum=jspectrum,
scoreMin=scoremin,
scoreMax=scoremax)
composite.add_view(view)
# make track
for sample, wfs in files_dic.iteritems():
for wf in wfs:
sampname = os.path.basename(wf)
bname = sampname.split(".")[0]
track = Track(name=bname,
tracktype=ftype,
url=os.path.join(url_base, "%s" % sampname),
local_fn=os.path.abspath(wf),
remote_fn=os.path.join(upload_base, "%s" % sampname),
visibility=jvis,
shortLabel=bname,
longLabel=bname,
spectrum=jspectrum,
scoreMin=scoremin,
scoreMax=scoremax,
subgroups={"sample": sample})
view.add_tracks(track)
trackdb.add_tracks(composite)
print('Track looks like this:')
print(trackdb)
if args.render:
# print('Rendering to %s' % hub.local_fn)
# results = hub.render()
# upload_hub(hub=hub, host='localhost', remote_dir='example_grouping_hub')
stage_hub(hub, staging=args.outdir)
if args.upload:
print('Uploading to [email protected]')
# for track in trackdb.tracks:
# upload_track(track = track, host = host, user = user)
upload_hub(hub=hub,
host=host,
user=user,
remote_dir="/data/web/sites/motevo_from_peaks")
print('Subgroups:')
for sg in subgroups:
print(sg)
print("Staging to path: %s" % args.outdir)
def composite_track(self):
"""
Create a composite track ready to be added to a trackhub.TrackDb
instance.
"""
bigwigs = [i for i in self.tracks if i.type_ == 'bigwig']
bigbeds = [i for i in self.tracks if i.type_ == 'bigbed']
# Build the HTML docs
last_section = self.reference_section()
html_string = utils.reST_to_html(
self.metadata['study'].get('description', '') + '\n' +
last_section)
sanitized_label = utils.sanitize(self.label, strict=True)
# Composite track to hold all subtracks for the study
composite = CompositeTrack(
name=sanitized_label,
short_label=self.study['short_label'],
long_label=self.study['long_label'],
tracktype='bigBed',
# Add all the documentation
html_string=html_string)
# If there are any bigWigs defined for this study, make a new "signal"
# subtrack in the composite and then add the bigWigs to it.
#
# Uses the sanitized label for the study to ensure uniqueness among
# tracks.
#
def _add_tracks(data_list, view, default_tracktype):
for data_obj in data_list:
kwargs = data_obj.obj.get('trackinfo', {})
kwargs = dict((k, str(v)) for k, v in kwargs.items())
kwargs.setdefault('tracktype', default_tracktype)
view.add_tracks(
Track(name=sanitized_label +
utils.sanitize(data_obj.label),
short_label=data_obj.label,
long_label=data_obj.obj['long_label'],
local_fn=data_obj.processed,
**kwargs))
if len(bigwigs) > 0:
signal_view = ViewTrack(
name=sanitized_label + 'signalviewtrack',
view=sanitized_label + 'signalview',
short_label=self.label + ' signal view',
long_label=self.label + ' signal view',
visibility='full',
maxHeightPixels='100:25:8',
autoScale='off',
tracktype='bigWig',
)
composite.add_view(signal_view)
_add_tracks(bigwigs, signal_view, 'bigWig')
# Same thing with bigBeds
if len(bigbeds) > 0:
bed_view = ViewTrack(
name=sanitized_label + 'bedviewtrack',
view=sanitized_label + 'bed_view',
short_label=self.label + ' bed view',
long_label=self.label + ' bed view',
visibility='dense',
)
composite.add_view(bed_view)
_add_tracks(bigbeds, bed_view, 'bigBed 9')
return composite
return '0,0,0'
hub, genomes_file, genome, trackdb = default_hub(**config['trackhub']['hub'])
hub.remote_fn = config['trackhub']['upload']['remote_fn']
sample_composite = CompositeTrack(name='samplespecific',
short_label='Samples',
long_label='Individual samples',
tracktype='bigWig')
sample_signal_view = ViewTrack(name='samplesignalview',
view='samplesignal',
visibility='full',
tracktype='bigWig',
viewLimits='0:150000',
maxHeightPixels='15:50:127',
short_label='Sample Signal',
long_label='Sample-level Signal')
sample_bed_view = ViewTrack(name='samplebedview',
view='samplebed',
visibility='dense',
tracktype='bigBed',
short_label='Sample Region',
long_label='Sample-level regions')
bait_composite = CompositeTrack(name='baitspecific',
short_label='Baits',
long_label='Bait level',
tracktype='bigBed')
dragAndDrop='subtracks',
visibility='full',
viewLimits='1:10',
autoScale='off',
maxHeightPixels='100:30:10',
priority=1
)
#######################
#make the tracks and add to the appropriate groups during loo
chipseq_bwfiles = [f for f in os.listdir(datadir+'/chip/processed/bw/') if f.endswith('.bw')]
chipseq_bbfiles = [b for b in os.listdir(datadir+'/chip/processed/bb/') if b.endswith('.bb')]
signal_view = ViewTrack(
name = 'Chipseq_signal',
view = 'Signal',
visibility = 'full',
tracktype = 'bigWig',
short_label = 'signal')
bed_view = ViewTrack(
name = 'Chipseq_bed',
view = 'Bed',
visibility = 'dense',
tracktype = 'bigBed',
short_label = 'peaks')
comp.add_view(signal_view)
comp.add_view(bed_view)
for f in chipseq_bwfiles:
name, rep = get_name_and_rep(f)
bname = os.path.basename(f)
short_label = 'signal')
<<<<<<< HEAD
bed_view = ViewTrack(
=======
chip_bed_view = ViewTrack(
>>>>>>> e33a5d89c0f5d75afae01dde9452063e2287f8d6
name = 'Chipseq_bed',
view = 'Bed',
visibility = 'dense',
tracktype = 'bigBed',
short_label = 'peaks')
<<<<<<< HEAD
rna_view = ViewTrack(
name = 'RNAseq_signal',
view = 'RNA',
visibility = 'full',
tracktype = 'bigWig',
short_label = 'Rnaseq_signal')
comp_chip.add_view(signal_view)
comp_chip.add_view(bed_view)
comp_rna.add_view(rna_view)
=======
#rna_signal_view = ViewTrack(
# name = 'RNAseq_signal',
# view = 'Signal',
# visibility = 'full',
# tracktype = 'bigwig',
# short_label = 'signal')
#
comp.add_view(chip_signal_view)
args.composite_track_name,
tracktype="bigWig")
# After the composite track has been created, we can incrementally add additional parameters.
# This is same method can be used for all classes derived from Track CompositeTrack, ViewTrack
# and of course Track itself:
composite.add_params(dragAndDrop='subtracks', visibility='full')
# The next part of the hierarchy is a ViewTrack object. Both ViewTrack and CompositeTrack are subclasses of the more generic Track class,
# so they act in much the same way. This should look familiar, but a notable difference is the addition of the view kwarg
from trackhub import ViewTrack
bed_view = ViewTrack(name="bedViewTrack",
view="Bed",
visibility="squish",
tracktype="bigBed 3",
short_label="beds",
long_label="Beds")
signal_view = ViewTrack(name="signalViewTrack",
view="Signal",
visibility="full",
tracktype="bigWig 0 10000",
short_label="signal",
long_label="Signal")
# Add these new view tracks to composite:
composite.add_view(bed_view)
composite.add_view(signal_view)
viewLimits='1:10',
autoScale='off',
maxHeightPixels='100:30:10',
priority=1)
#######################
#make the tracks and add to the appropriate groups during loo
chipseq_bwfiles = [
f for f in os.listdir(datadir + '/chip/processed/bw/') if f.endswith('.bw')
]
chipseq_bbfiles = [
b for b in os.listdir(datadir + '/chip/processed/bb/') if b.endswith('.bb')
]
signal_view = ViewTrack(name='Chipseq_signal',
view='Signal',
visibility='full',
tracktype='bigWig',
short_label='signal')
bed_view = ViewTrack(name='Chipseq_bed',
view='Bed',
visibility='dense',
tracktype='bigBed',
short_label='peaks')
comp.add_view(signal_view)
comp.add_view(bed_view)
for f in chipseq_bwfiles:
name, rep = get_name_and_rep(f)
bname = os.path.basename(f)
col = maps[name]
track = Track(name='%s_%s_signal' % (name, rep),