def make_bigwig_trackhub(libraries, trackdb):
cluster = trackhub.SubGroupDefinition(
name='cluster',
label='cluster',
mapping=cluster_mapping)
subgroups = [
cluster,
trackhub.SubGroupDefinition(
name='multi',
label='multi',
mapping={
'uniq': 'uniq',
'all': 'all',
}),
]
composite = trackhub.CompositeTrack(
name='composite',
short_label='signal',
dimensions='dimX=cluster dimY=multi',
tracktype='bigWig',
visibility='dense',
)
composite.add_subgroups(subgroups)
trackdb.add_tracks(composite)
signal_view = trackhub.ViewTrack(
name='signalviewtrack',
view='signal',
visibility='dense',
tracktype='bigWig',
short_label='Signal')
composite.add_view(signal_view)
for i, (library_id, row) in enumerate(libraries.iterrows()):
for track_index, track_type in enumerate(['uniq', 'all']):
cluster_assignment = human_cluster_mapping[row.cluster_assignment]
priority = i * 10 + track_index
track = trackhub.Track(
url=make_home_url(row[track_type]),
name="{:04d}".format(priority) + '_' + row.analysis_name + '_' + track_type,
visibility='full',
tracktype='bigWig',
subgroups={'cluster': cluster_assignment, 'multi': track_type},
color=hex_to_ucsc_color(colors[row.cluster_assignment]),
priority=priority,
)
signal_view.add_tracks(track)
def make_bam_trackhub(libraries, trackdb):
cluster = trackhub.SubGroupDefinition(
name='cluster',
label='cluster',
mapping=cluster_mapping)
bam_composite = trackhub.CompositeTrack(
name='reads',
short_label='reads',
dimensions='dimX=cluster',
tracktype='bam',
visibility='dense',
)
bam_composite.add_subgroups([cluster])
trackdb.add_tracks(bam_composite)
bam_view = trackhub.ViewTrack(
name='readview',
view='reads',
visibility='dense',
tracktype='bam',
short_label='Reads')
bam_composite.add_view(bam_view)
curdir = os.getcwd()
#os.chdir(public_dir)
try:
for i, (library_id, row) in enumerate(libraries.iterrows()):
cluster_assignment = human_cluster_mapping[row.cluster_assignment]
priority = "{:03d}".format(i)
path, name = os.path.split(row.bam)
bam_index = row.bam + '.bai'
if not os.path.exists(name):
os.symlink(row.bam, name)
if not os.path.exists(name + '.bai'):
os.symlink(bam_index, name + '.bai')
track = trackhub.Track(
url=make_home_url(row['bam']),
name=priority + '_' + row.analysis_name + '_reads',
visibility='full',
tracktype='bam',
subgroups={'cluster': cluster_assignment},
color=hex_to_ucsc_color(colors[row.cluster_assignment]),
priority=i,
)
bam_view.add_tracks(track)
finally:
os.chdir(curdir)
def create_trackhub(OutFolder,ListFile,Genome,ChrSize, EMAIL,DesignFile):
ERROR_STR = 'ERROR: Please check design file'
HEADER = ['group', 'replicate', 'fastq_1', 'fastq_2']
makedir(OutFolder)
dIn = open(DesignFile, 'r')
header = dIn.readline().strip().split(',')
if header[:4] != HEADER:
print("{} header: {} != {}".format(ERROR_STR,','.join(header),','.join(HEADER)))
sys.exit(1)
paramColn = {}
for i in range(len(header)):
if header[i][:6]=="track_": # header start with track_
paramColn[header[i][6:]]=i
sampleDesignDict = {}
designDict = {}
if paramColn:
while True:
line = dIn.readline()
if line:
lspl = [x.strip() for x in line.strip().split(',')]
lspl[0] = [lspl[0]+Postfix[1], lspl[0]+'_R'+lspl[1]]
lspl[0] = [trackhub.helpers.sanitize(lspl[0][0].replace(".", "_"), strict=False),trackhub.helpers.sanitize(lspl[0][1].replace(".", "_"), strict=False)]
sampleDesignDict[lspl[0][0]] = {}
sampleDesignDict[lspl[0][1]] = {}
for k in paramColn.keys():
sampleDesignDict[lspl[0][0]][k]=lspl[paramColn[k]]
sampleDesignDict[lspl[0][1]][k]=lspl[paramColn[k]]
if k in designDict:
designDict[k][lspl[paramColn[k]]] = lspl[paramColn[k]]
else:
designDict[k] = {lspl[paramColn[k]]:lspl[paramColn[k]]}
else:
break
dIn.close()
fileList = []
fin = open(ListFile,'r')
while True:
line = fin.readline()
if line:
ifile = [x.strip() for x in line.strip().split('\t')]
colour = ""
if sampleDesignDict:
kfile = trackhub.helpers.sanitize(ifile[0].replace(".", "_"), strict=False)
if kfile in sampleDesignDict:
if "color" in sampleDesignDict[kfile]:
h = sampleDesignDict[kfile]["color"].lstrip('#')
colour = ','.join(str(x) for x in tuple(int(h[i:i+2], 16) for i in (0, 2, 4)))
if len(colour.strip()) == 0:
colour = '0,0,178'
fileList.append((ifile[1],colour,ifile[0]))
else:
break
fin.close()
fileList = sorted(fileList, key=lambda x: x[2])
# Initialize the components of a track hub, already connected together
hub, genomes_file, genome, trackdb = trackhub.default_hub(
hub_name="RNISRS_hub",
short_label='Regeneromics Shared Resource hub',
long_label='Regeneration Next Initiative Regeneromics Shared Resource hub',
genome=Genome,
email=EMAIL)
# create compositeTracks
if sampleDesignDict:
composite = trackhub.CompositeTrack(
name = 'composite',
short_label='singlal'
)
# Add those subgroups to the composite track
subgroups = []
for k in designDict.keys():
if k!='color':
subg = trackhub.SubGroupDefinition(
name=k,
label=k,
mapping=designDict[k]
)
subgroups.append(subg)
composite.add_subgroups(subgroups)
# Add the composite track to the trackDb
trackdb.add_tracks(composite)
signal_view = trackhub.ViewTrack(
name='signalviewtrack',
view='signal',
short_label='Signal')
composite.add_view(signal_view)
regions_view = trackhub.ViewTrack(
name='regionsviewtrack',
view='regions',
short_label='Regions')
composite.add_view(regions_view)
for ifile,color,id in fileList:
extension = os.path.splitext(ifile)[1].replace(".", "").lower()
filename = trackhub.helpers.sanitize(os.path.splitext(os.path.basename(ifile))[0].replace(".", "_"), strict=False)
if extension in ['bed','broadpeak','narrowpeak']:
# convert bed to bigbed
# sort bed file
cmd = "sort -k1,1 -k2,2n " + ifile +" >" + ifile + "_sorted.bed"
os.system(cmd)
# bedToBigBed
os.system("awk '$1 != \"track\" {$5=($5>1000)?1000:$5; print ;}' "+ifile+"_sorted.bed > "+ifile+"_srt.bed")
if extension == "bed":
cmd = "bedToBigBed "+ifile+"_srt.bed"+" "+ChrSize+" "+ifile+".bb"
extension = "bb"
if extension == "broadpeak":
cmd = "bedToBigBed -type=bed6+3 -as=broadPeak.as "+ifile+"_srt.bed"+" "+ChrSize+" "+ifile+".bb"
if extension == "narrowpeak":
cmd = "bedToBigBed -type=bed6+4 -as=narrowPeak.as "+ifile+"_srt.bed"+" "+ChrSize+" "+ifile+".bb"
os.system(cmd)
# change ifile to new bigbed file
ifile = ifile+".bb"
if extension in TrackType.keys():
if sampleDesignDict:
track = trackhub.Track(
name=filename,
source=filename,
short_label=id,
long_label=filename,
color=color,
visibility=Visibility[extension],
tracktype=TrackType[extension],
subgroups=sampleDesignDict[filename],
autoScale='on')
signal_view.add_tracks(track)
else:
track = trackhub.Track(
name=filename,
source=filename,
short_label=id,
long_label=filename,
color=color,
visibility=Visibility[extension],
tracktype=TrackType[extension],
autoScale='on')
trackdb.add_tracks(track)
linkname=os.path.join(OutFolder, Genome, filename+"."+TrackType[extension].split()[0])
makedir(os.path.join(OutFolder, Genome))
os.symlink("../../"+ifile, linkname)
else:
pass
hub.render(staging=OutFolder)
def make_trackhub():
import trackhub
data = read_peng_20180710_cluster_memberships()
df = pandas.DataFrame(
data, columns=['cell_id', 'cluster_name', 'value', 'color'])
df['rgb'] = df['color'].apply(
lambda channel: ','.join([str(int(x * 255)) for x in channel]))
colors = df[['cluster_name', 'value', 'color', 'rgb']].drop_duplicates()
hub, genomes_file, genome, trackdb = trackhub.default_hub(
hub_name="C1_peng_20180710_cluster",
short_label="C1_peng_20180710_cluster",
genome="mm10",
email="*****@*****.**")
subgroups = [
trackhub.SubGroupDefinition(name='multiread',
label='multiread',
mapping={
'all': 'all_reads',
'uniq': 'unique_only',
})
]
composite = trackhub.CompositeTrack(
name='composite',
short_label='bigwigs',
dimensions='dimX=multiread',
sortOrder='multiread',
visibility='full',
tracktype='bigWig',
)
composite.add_subgroups(subgroups)
trackdb.add_tracks(composite)
signal_view = trackhub.ViewTrack(name='signal',
view='signal',
visibility='full',
tracktype='bigWig')
composite.add_view(signal_view)
subgroup_map = {
'-mm10-M4-male_all.bw': 'all',
'-mm10-M4-male_uniq.bw': 'uniq'
}
for suffix in ['-mm10-M4-male_all.bw', '-mm10-M4-male_uniq.bw']:
for i, row in colors.iterrows():
name = row.value + subgroup_map[suffix]
url = 'http://woldlab.caltech.edu/~diane/C1_peng_20180710_cluster_bigwigs/' + row.value + suffix
#url = '../' + row.value + suffix
track = trackhub.Track(
name=trackhub.helpers.sanitize(name),
long_label=row.cluster_name + ' ' + subgroup_map[suffix],
url=url,
#source=os.path.join('C1_peng_20180710_cluster_bigwigs/', name),
visibility='full',
tracktype='bigWig',
subgroups={'multiread': subgroup_map[suffix]},
color=row.rgb)
signal_view.add_tracks(track)
print(trackdb)
trackhub.upload.upload_hub(
hub=hub,
host='localhost',
remote_dir=
'/woldlab/loxcyc/home/diane/public_html/C1_peng_20180710_cluster_bigwigs/'
)
email="*****@*****.**")
# Subgroups provide a way of tagging tracks. If you give the same tag to
# a bigwig and a bigbed, you can use that tag to select them both. This is
# useful for, e.g., ChIP-seq tracks where the tag is "sample" or
# "antibody", and you can select both the signal and the called peaks at
# the same time.
#
# Here we're making some contrived subgroups just to illustrate their use.
subgroups = [
# A subgroup to select the replicate number
trackhub.SubGroupDefinition(name='num',
label='Number',
mapping={
'0': 'zero',
'1': 'one',
'2': 'two',
}),
# A contrived subgroup that will only tag it as "yes" if it's sample 1.
trackhub.SubGroupDefinition(name='s1',
label='is_sample_1',
mapping={
'y': 'Yes',
'n': 'No',
}),
# While the two different views we create below are a good way of
# turning on/off the signal or regions in bulk, this subgroup allows us
# to sort the tracks by "name" and then by "kind". This is helpful for
def create_trackhub(OutFolder,ListFile,Genome,EMAIL,DesignFile,Postfix,PathPrefix=''):
ERROR_STR = 'ERROR: Please check design file'
HEADER = ['group', 'replicate', 'fastq_1', 'fastq_2']
makedir(OutFolder)
dIn = open(DesignFile, 'r')
header = dIn.readline().strip().split(',')
if header[:4] != HEADER:
print("{} header: {} != {}".format(ERROR_STR,','.join(header),','.join(HEADER)))
sys.exit(1)
paramColn = {}
for i in range(len(header)):
if header[i][:6]=="track_":
paramColn[header[i][6:]]=i
sampleDesignDict = {}
designDict = {}
if paramColn:
while True:
line = dIn.readline()
if line:
lspl = [x.strip() for x in line.strip().split(',')]
lspl[0] = [lspl[0]+Postfix[1], lspl[0]+'_R'+lspl[1]+Postfix[0]]
lspl[0] = [trackhub.helpers.sanitize(lspl[0][0].replace(".", "_"), strict=False),trackhub.helpers.sanitize(lspl[0][1].replace(".", "_"), strict=False)]
sampleDesignDict[lspl[0][0]] = {}
sampleDesignDict[lspl[0][1]] = {}
for k in paramColn.keys():
sampleDesignDict[lspl[0][0]][k]=lspl[paramColn[k]]
sampleDesignDict[lspl[0][1]][k]=lspl[paramColn[k]]
if k in designDict:
designDict[k][lspl[paramColn[k]]] = lspl[paramColn[k]]
else:
designDict[k] = {lspl[paramColn[k]]:lspl[paramColn[k]]}
else:
break
dIn.close()
fileList = []
fin = open(ListFile,'r')
while True:
line = fin.readline()
if line:
ifile,colour = line.strip().split('\t')
if sampleDesignDict:
kfile = trackhub.helpers.sanitize(os.path.splitext(os.path.basename(ifile))[0].replace(".", "_"), strict=False)
if kfile in sampleDesignDict:
if "color" in sampleDesignDict[kfile]:
h = sampleDesignDict[kfile]["color"].lstrip('#')
colour = ','.join(str(x) for x in tuple(int(h[i:i+2], 16) for i in (0, 2, 4)))
if len(colour.strip()) == 0:
colour = '0,0,178'
fileList.append((PathPrefix.strip()+ifile,colour))
else:
break
fin.close()
# Initialize the components of a track hub, already connected together
hub, genomes_file, genome, trackdb = trackhub.default_hub(
hub_name="RNISRS_hub",
short_label='Regeneromics Shared Resource hub',
long_label='Regeneration Next Initiative Regeneromics Shared Resource hub',
genome=Genome,
email=EMAIL)
# create compositeTracks
if sampleDesignDict:
composite = trackhub.CompositeTrack(
name = 'composite',
short_label='singlal'
)
# Add those subgroups to the composite track
subgroups = []
for k in designDict.keys():
if k!='color':
subg = trackhub.SubGroupDefinition(
name=k,
label=k,
mapping=designDict[k]
)
subgroups.append(subg)
composite.add_subgroups(subgroups)
# Add the composite track to the trackDb
trackdb.add_tracks(composite)
signal_view = trackhub.ViewTrack(
name='signalviewtrack',
view='signal',
short_label='Signal')
composite.add_view(signal_view)
regions_view = trackhub.ViewTrack(
name='regionsviewtrack',
view='regions',
short_label='Regions')
composite.add_view(regions_view)
for ifile,color in fileList:
extension = os.path.splitext(ifile)[1].replace(".", "").lower()
filename = trackhub.helpers.sanitize(os.path.splitext(os.path.basename(ifile))[0].replace(".", "_"), strict=False)
if extension in ['bed','broadpeak','narrowpeak']:
pass
elif extension in TrackType.keys():
if filename in sampleDesignDict:
track = trackhub.Track(
name=filename,
source=ifile,
color=color,
visibility=Visibility[extension],
tracktype=TrackType[extension],
subgroups=sampleDesignDict[filename],
autoScale='on')
signal_view.add_tracks(track)
else:
track = trackhub.Track(
name=filename,
source=ifile,
color=color,
visibility=Visibility[extension],
tracktype=TrackType[extension],
autoScale='on')
trackdb.add_tracks(track)
linkname=os.path.join(OutFolder, Genome, filename+"."+TrackType[extension])
makedir(os.path.join(OutFolder, Genome))
os.symlink(ifile, linkname)
else:
pass
hub.render(staging=OutFolder)