def setup(self):
self.hub = Hub(hub='example_hub',
short_label='example hub',
long_label='an example hub for testing',
email='*****@*****.**')
self.genomes_file = GenomesFile()
self.genome = Genome('dm3')
self.trackdb = TrackDb()
self.tracks = [
Track(name='track1',
tracktype='bigBed',
source=os.path.join(d, 'random-hg38-0.bigBed')),
Track(
name='track2',
tracktype='bigWig',
source=os.path.join(d, 'sine-hg38-0.bedgraph.bw'),
),
Track(
name='track3',
tracktype='bigWig',
source=os.path.join(d, 'sine-hg38-1.bedgraph.bw'),
filename='3.bw',
)
]
self.hub.add_genomes_file(self.genomes_file)
self.genomes_file.add_genome(self.genome)
self.genome.add_trackdb(self.trackdb)
self.trackdb.add_tracks(self.tracks)
def setup(self):
with pytest.warns(DeprecationWarning):
self.hub = Hub(
hub='example_hub',
short_label='example hub',
long_label='an example hub for testing',
email='*****@*****.**')
self.genomes_file = GenomesFile()
self.genome = Genome('dm3')
self.trackdb = TrackDb()
self.tracks = [
Track(
name='track1',
tracktype='bigBed',
local_fn=os.path.join(d, 'random-hg38-0.bigBed'),
remote_fn='1.bigbed',
),
Track(
name='track2',
tracktype='bigWig',
local_fn=os.path.join(d, 'sine-hg38-0.bedgraph.bw'),
remote_fn='2.bw',
),
]
self.hub.add_genomes_file(self.genomes_file)
self.genomes_file.add_genome(self.genome)
self.genome.add_trackdb(self.trackdb)
self.trackdb.add_tracks(self.tracks)
def create_hub(geo,
email=None,
upload=False,
upload_dir=".",
user=None,
host=None):
""" Create an UCSC track hub from a Geo object
"""
hub = Hub(hub=geo.gse,
short_label=geo.gse,
long_label="Hub for {0}".format(geo.gse),
email=email)
genomes_file = GenomesFile()
trackdb = TrackDb()
local_dir = geo.gse
#hub.remote_fn = os.path.join(upload_dir, geo.gse, os.path.basename(hub.local_fn))
all_tracks = {}
for sample in geo.samples.values():
genome = sample['genome']
all_tracks.setdefault(genome, [])
name = re.sub('[^0-9a-zA-Z]+', '_', sample['name'])
track = Track(
name=name,
url=os.path.join(HUB_URLBASE, geo.gse, genome,
"{0}.bw".format(sample['gsm'])),
tracktype='bigWig',
short_label=sample['gsm'],
long_label=name,
color='128,128,0',
maxHeightPixels='30:30:11',
)
basename = os.path.basename(track.url)
track.local_fn = os.path.join(local_dir, basename)
track.remote_fn = os.path.join(upload_dir, geo.gse, genome, basename)
all_tracks[genome].append(track)
for build, tracks in all_tracks.items():
genome = Genome(build)
trackdb.add_tracks(tracks)
genome.add_trackdb(trackdb)
genomes_file.add_genome(genome)
hub.add_genomes_file(genomes_file)
results = hub.render()
if upload:
for track in trackdb.tracks:
upload_track(track=track, host=host, user=user)
upload_hub(hub=hub, host=host, user=user)
def setup(self):
self.hub = Hub(hub='example_hub',
short_label='example hub',
long_label='an example hub for testing',
email='*****@*****.**')
self.genomes_file = GenomesFile()
self.genome = Genome('dm3')
self.trackdb = TrackDb()
self.tracks = [
Track(name='track1', tracktype='bam'),
Track(name='track2', tracktype='bigWig'),
]
def setup(self):
self.hub = Hub(hub='example_hub',
short_label='example hub',
long_label='an example hub for testing',
email='*****@*****.**')
self.genomes_file = GenomesFile()
self.genome = Genome('dm3')
self.trackdb = TrackDb()
self.tracks = [
Track(name='track1', tracktype='bam', local_fn='data/track1.bam'),
Track(
name='track2',
tracktype='bigWig',
local_fn='data/track2.bigwig',
),
]
self.hub.add_genomes_file(self.genomes_file)
self.genomes_file.add_genome(self.genome)
self.genome.add_trackdb(self.trackdb)
self.trackdb.add_tracks(self.tracks)
if not args.long_label:
args.long_label = args.short_label
upload_dir = os.path.join(args.upload_dir, args.hub)
#THIS IS REALLY BAD AND NOT INUITITVE
if args.no_s3:
URLBASE = os.path.join(
"https://s3-us-west-1.amazonaws.com/sauron-yeo/", args.hub)
else:
URLBASE = os.path.join("http://sauron.ucsd.edu/Hubs", args.hub)
GENOME = args.genome
hub = Hub(
hub=args.hub,
short_label=args.short_label,
long_label=args.long_label,
email=args.email,
)
genomes_file = GenomesFile()
genome = Genome(GENOME)
trackdb = TrackDb()
supertrack = SuperTrack(name=args.hub,
short_label=args.hub,
long_label=args.hub)
genome.add_trackdb(trackdb)
genomes_file.add_genome(genome)
hub.add_genomes_file(genomes_file)
hub.upload_fn = upload_dir
files = args.files
def main():
parser = argparse.ArgumentParser(
description=
'Takes in files to turn into trackhub. This version automatically ')
# tracks files
##############
parser.add_argument('files',
nargs='+',
help='Files to turn into track hub')
# namings
#########
parser.add_argument('--hub', help="hub name (no spaces)", required=True)
parser.add_argument('--genome', help="genome name", required=True)
# upload (in fact run_local=True)
########
#parser.add_argument('--no_s3', default=False, action="store_true", help="upload to defined server instead of s3")
#parser.add_argument('--serverscp', default="tscc-login2.sdsc.edu", help="server to SCP to")
#parser.add_argument('--user', default='adomissy', help="that is uploading files")
# parser.add_argument('--uploaddir', default='yeolab-trackhubs', help="directory to upload files to if not uploading to aws")
# web access
############
# parser.add_argument('--urldomain', default="s3-us-west-2.amazonaws.com", help="url domain for public access to trackhubs")
# parser.add_argument('--urldir', default="yeolab-trackhubs", help="url directory for public access to trackhubs")
# hub labels
############
# parser.add_argument('--hub_short_label', default=None, help="short label for hub")
# parser.add_argument('--hub_long_label', default=None, help="long label for hub")
parser.add_argument('--hub_email',
default='*****@*****.**',
help="email for hub")
# name parts grouping
#####################
parser.add_argument('--sep', default=".", help="Seperator")
parser.add_argument('--num_sep',
default=2,
type=int,
help="Number of seperators deep to group on")
###########################################################################
args = parser.parse_args()
# TODO: unhack this, but let's keep all trackhubs here for now.
urldomain = "s3-us-west-2.amazonaws.com"
urldir = "yeolab-trackhubs"
uploaddir = "yeolab-trackhubs"
hub_name = args.hub
hub_email = args.hub_email
# default label settings
########################
hub_short_label = hub_name
hub_long_label = hub_name
# hard coding serverscp, in variable HOST
HOST = "localhost"
# hard coding user, in variable USER
USER = "******"
GENOME = args.genome
# hack for tutorial dataset so it is easy to view in ucsd genome browser
if GENOME == 'hg19chr19kbp255':
GENOME == 'hg19'
uploaddir = os.path.join(uploaddir, hub_name)
URLBASE = os.path.join("http://" + urldomain + "/" + urldir + "/",
hub_name)
# create data structures
########################
hub = Hub(
hub=hub_name,
short_label=hub_short_label,
long_label=hub_long_label,
email=hub_email,
)
hub.upload_fn = uploaddir
genomes_file = GenomesFile()
hub.add_genomes_file(genomes_file)
genome = Genome(GENOME)
genomes_file.add_genome(genome)
trackdb = TrackDb()
genome.add_trackdb(trackdb)
supertrack = SuperTrack(name=hub_name,
short_label=hub_short_label,
long_label=hub_long_label)
# separate bigwigs, bigbeds and others for different processing methods
#######################################################################
bigwig_files = [
file for file in args.files
if file.endswith(".posbw") or file.endswith(".negbw") or file.endswith(
".bw") or file.endswith(".bigWig") or file.endswith(".bigwig")
]
bigbed_files = [
file for file in args.files if file.endswith(".bb")
or file.endswith(".bigBed") or file.endswith(".bigbed")
]
# not used
#other_files = [file for file in args.files if (file not in bigwig_files and file not in bigbed_files )]
# process bigwig files , re-grouped by third 2 dot-sepatarated name-parts, as multitracks
##########################################################################################
key_func = lambda x: x.split(args.sep)[:args.num_sep]
for group_key, group_bigwig_files in groupby(
sorted(bigwig_files, key=key_func), key_func):
group_bigwig_files_list = list(group_bigwig_files)
print("args sep: {}".format(args.sep))
print("args num sep: {}".format(args.num_sep))
print("split filename: {}".format(bigwig_files[0].split(
args.sep)[:args.num_sep]))
print "-----------------------------------------"
print "processing bigwig files group with key :", group_key
print "comprised of following files:", group_bigwig_files_list
print "-----------------------------------------"
long_name = remove_plus_and_pct(
os.path.basename(args.sep.join(group_key[:args.num_sep])))
aggregate = AggregateTrack(name=long_name,
tracktype='bigWig',
short_label=long_name,
long_label=long_name,
aggregate='transparentOverlay',
showSubtrackColorOnUi='on',
autoScale='on',
priority='1.4',
alwaysZero='on',
visibility="full")
for bigwigfile in group_bigwig_files_list:
print "--------------------------"
print "bigwigfile", bigwigfile
print "--------------------------"
base_track = remove_plus_and_pct(os.path.basename(bigwigfile))
split_track = base_track.split(args.sep)
long_name = args.sep.join(split_track[:args.num_sep] +
split_track[-3:])
color = "0,100,0" if "pos" in bigwigfile else "100,0,0"
track = Track(name=long_name,
url=os.path.join(URLBASE, GENOME, base_track),
tracktype="bigWig",
short_label=long_name,
long_label=long_name,
color=color,
local_fn=bigwigfile,
remote_fn=os.path.join(uploaddir, GENOME,
base_track))
#print "aggregate.add_subtrack", track.name
aggregate.add_subtrack(track)
#print "supertrack.add_track", aggregate
supertrack.add_track(aggregate)
#print "trackdb.add_tracks", aggregate
#trackdb.add_tracks(aggregate)
# process bigbed files as single track
######################################
for bigbed_file in bigbed_files:
# print "--------------------------"
# print "bigbedfile", bigbedfile
# print "--------------------------"
color = "0,100,0" if "pos" in bigbed_file else "100,0,0"
base_track = remove_plus_and_pct(os.path.basename(bigbed_file))
long_name = args.sep.join(base_track.split(
args.sep)[:args.num_sep]) + ".bb"
track = Track(name=long_name,
url=os.path.join(URLBASE, GENOME, base_track),
tracktype="bigBed",
short_label=long_name,
long_label=long_name,
color=color,
local_fn=bigbed_file,
remote_fn=os.path.join(uploaddir, GENOME, base_track),
visibility="full")
#trackdb.add_tracks(track)
supertrack.add_track(track)
trackdb.add_tracks(supertrack)
result = hub.render()
hub.remote_fn = os.path.join(uploaddir, "hub.txt")
# process bigbed files (bam?)
######################
## UNUSED
# if bigwigfile.endswith(".bw") or bigwigfile.endswith('.bigWig'): tracktype = "bigWig"
# if bigwigfile.endswith(".bb") or bigwigfile.endswith('.bigBed'): tracktype = "bigBed"
# if bigwigfile.endswith(".bam"): tracktype = "bam"
# 'upolading' (locally)
########################
for track in trackdb.tracks:
#print("upload_track(track=" + track.__repr__() + ", host=" + args.serverscp + ", user="******"run_local=True")
#upload_track(track=track, host=args.serverscp, user=args.user)
# upload_track(track=track, host=args.serverscp, user=args.user, run_s3=args.no_s3)
upload_track(track=track, host=HOST, user=USER, run_local=True)
#print("upload_hub(hub=" + hub.__repr__() + ", host=" + args.serverscp + ", user="******"run_local=True")
#upload_hub(hub=hub, host=args.serverscp, user=args.user)
# upload_hub(hub=hub, host=args.serverscp, user=args.user, run_s3=args.no_s3)
pass
upload_hub(hub=hub, host=HOST, user=USER, run_local=True)
#
print("UPLOADDIR: {}".format(uploaddir))
print("BUCKET: {}".format(uploaddir))
copy_dir_to_aws(
src=uploaddir,
dest=uploaddir,
)
print("FINAL URL: {}/hub.txt".format(URLBASE))
def files2viz(files_to_visualize):
if DEBUG:
print "Visualizing"
print files_to_visualize
from trackhub import Hub, GenomesFile, Genome, TrackDb, Track
from trackhub.upload import upload_hub
HUBHOST = 'http://cherry-vm45.stanford.edu'
HUBDIR = 'trackhubs'
USER = '******'
URLBASE = os.path.join(HUBHOST, HUBDIR)
EDWBASE = 'http://encodedcc.sdsc.edu/warehouse'
GENOME = 'hg19'
hub = Hub(hub='Selected_ENCODE_Tracks',
short_label='Selected_ENCODE_Tracks_short',
long_label='Selected_ENCODE_Tracks_long',
email='*****@*****.**')
genomes_file = GenomesFile()
genome = Genome(GENOME)
trackdb = TrackDb()
for accession in files_to_visualize:
file_obj = get_ENCODE(accession)
if DEBUG:
print file_obj
if file_obj['file_format'] == 'bigWig':
track = Track(name=accession,
url=os.path.join(EDWBASE,
str(file_obj['download_path'])),
tracktype='bigWig',
short_label=accession,
long_label=accession,
color='128,0,0',
visibility='full')
trackdb.add_tracks([track])
genome.add_trackdb(trackdb)
genomes_file.add_genome(genome)
hub.add_genomes_file(genomes_file)
results = hub.render()
if DEBUG:
print hub
print '...'
print genomes_file
print '...'
print genome
print '...'
print trackdb
#upload_hub(hub=hub, host=HUBHOST, user=USER) #doesn't seem to work
import subprocess
subprocess.call(
"cd .. && rsync -r trackhub [email protected]:/www/html/trackhubs",
shell=True)
import webbrowser
hubfile = str(hub.hub) + '.hub.txt'
UCSC_url = 'http://genome.ucsc.edu/cgi-bin/hgTracks?udcTimeout=1&db=hg19' + \
'&hubUrl=' + os.path.join(HUBHOST,HUBDIR,'trackhub',hubfile) # + \
#'&hsS_doLoadUrl=submit' + '&hgS_loadUrlName=' + os.path.join(HUBHOST,HUBDIR,'trackhub','session.txt')
print UCSC_url
webbrowser.open(UCSC_url)
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)
parser.add_argument("--email", help="Required. your email to contact")
parser.add_argument("--composite_track_name",
help="Required. the name of your composite track")
args = parser.parse_args()
assert args.hub_name is not None, "please provide the hub_name"
assert args.base_url is not None, "please provide the base_url"
assert args.composite_track_name is not None, "please provide the composite track name"
assert args.email is not None, "please provide your email"
assert args.input_dir is not None, "please provide the path to the bigwig and bigbed files on your local computer"
from trackhub import Hub, GenomesFile, Genome, TrackDb
hub = Hub(hub='%s' % args.hub_name,
short_label='%s' % args.hub_name,
long_label='%s ChIP-seq hub' % args.hub_name,
email='%s' % args.email)
genomes_file = GenomesFile()
genome = Genome('hg19')
trackdb = TrackDb()
# Bottom-up
genome.add_trackdb(trackdb)
genomes_file.add_genome(genome)
hub.add_genomes_file(genomes_file)
# make a composite track
from trackhub import CompositeTrack
composite = CompositeTrack(name="%s" % args.composite_track_name,
def ENCSR2viz(experiments_to_visualize):
if DEBUG:
print "Visualizing"
print experiments_to_visualize
from trackhub import Hub, GenomesFile, Genome, TrackDb, Track
from trackhub.upload import upload_hub
HUBHOST = 'http://cherry-vm45.stanford.edu'
HUBDIR = 'jseth/trackhubs'
USER = '******'
URLBASE = os.path.join(HUBHOST, HUBDIR)
EDWBASE = 'http://encodedcc.sdsc.edu/warehouse'
GENOME = 'hg19'
hub = Hub(
hub='ENCODE',
short_label='ENCODE',
long_label='ENCODE',
email='*****@*****.**')
genomes_file = GenomesFile()
genome = Genome(GENOME)
trackdb = TrackDb()
for accession in experiments_to_visualize:
experiment_obj = get_ENCODE(accession)
if DEBUG:
print experiment_obj
for file_id in experiment_obj['files']:
file_obj = get_ENCODE(file_id)
if file_obj['file_format'] in ['bigWig', 'bigBed', 'broadPeak', 'narrowPeak']:
if file_obj['file_format'] in ['bigWig']:
track_type = 'bigWig'
elif file_obj['file_format'] in ['bigBed', 'broadPeak', 'narrowPeak']:
track_type = 'bigBed'
track = Track(
name=str(file_obj['accession']),
url=os.path.join(EDWBASE, str(file_obj['download_path'])),
tracktype=track_type,
long_label=str(file_obj['accession']),
short_label=str(file_obj['output_type']),
color='128,0,0',
visibility='dense',
metadata='cell_type=primary')
print file_obj['accession']
trackdb.add_tracks([track])
genome.add_trackdb(trackdb)
genomes_file.add_genome(genome)
hub.add_genomes_file(genomes_file)
results=hub.render()
if DEBUG:
print hub
print '...'
print genomes_file
print '...'
print genome
print '...'
print trackdb
#upload_hub(hub=hub, host=HUBHOST, user=USER) #doesn't seem to work
import subprocess
subprocess.call("cd .. && rsync -r trackhub [email protected]:/www/html/jseth/trackhubs", shell=True)
import webbrowser
hubfile = str(hub.hub) + '.hub.txt'
UCSC_url = 'http://genome.ucsc.edu/cgi-bin/hgTracks?udcTimeout=1&db=hg19' + \
'&hubUrl=' + os.path.join(HUBHOST,HUBDIR,'trackhub',hubfile) # + \
#'&hsS_doLoadUrl=submit&hgS_loadUrlName=' + os.path.join(HUBHOST,HUBDIR,'trackhub','session.txt')
print UCSC_url
webbrowser.open(UCSC_url)
