def importRepeatsFromUCSC(infile, outfile, ucsc_database, repeattypes, genome):
'''import repeats from a UCSC formatted file.
The repeats are stored as a :term:`gff` formatted file.
'''
repclasses = "','".join(repeattypes.split(","))
# Repeats are either stored in a single ``rmsk`` table (hg19) or in
# individual ``rmsk`` tables (mm9) like chr1_rmsk, chr2_rmsk, ....
# In order to do a single statement, the ucsc mysql database is
# queried for tables that end in rmsk.
dbhandle = PipelineUCSC.connectToUCSC(
host=PARAMS["ucsc_host"],
user=PARAMS["ucsc_user"],
database=ucsc_database)
cc = dbhandle.execute("SHOW TABLES LIKE '%%rmsk'")
tables = [x[0] for x in cc.fetchall()]
if len(tables) == 0:
raise ValueError("could not find any `rmsk` tables")
tmpfile = P.getTempFile(shared=True)
total_repeats = 0
for table in tables:
E.info("%s: loading repeats from %s" % (ucsc_database, table))
cc = dbhandle.execute(
"""SELECT genoName, 'repeat', 'exon', genoStart+1, genoEnd, '.',
strand, '.',
CONCAT('class \\"', repClass, '\\"; family \\"', repFamily, '\\";')
FROM %(table)s
WHERE repClass in ('%(repclasses)s') """ % locals())
n = 0
for data in cc.fetchall():
n += 1
tmpfile.write("\t".join(map(str, data)) + "\n")
E.info("%s: %s=%i repeats downloaded" % (ucsc_database, table, n))
total_repeats += n
if total_repeats == 0:
raise ValueErrror("did not find any repeats for %s" % ucsc_database)
tmpfile.close()
tmpfilename = tmpfile.name
statement = '''cat %(tmpfilename)s
| %(pipeline_scriptsdir)s/gff_sort pos
| cgat gff2gff
--method=sanitize
--sanitize-method=genome
--skip-missing
--genome-file=%(genome)s
--log=%(outfile)s.log
| gzip
> %(outfile)s
'''
P.run()
os.unlink(tmpfilename)
def publish_tracks(export_files,
prefix="",
project_id=None,
project_name=None,
UCSC_ini=None):
'''publish a UCSC Track Hub.
This method takes a dictionary of file types associated
with files. For each file, a link will be created in
the upload directory. The track will be stored under
a project name, which will be derived from the location
of the working directory.
Information about the genome, the upload directory, etc. will be
taken from the global configuration dictionary.
For example, calling the following code in a pipeline executed
in .../proj013/mapping::
export_files = {
"bamfiles": glob.glob("*/*.bam") + glob.glob("*/*.bam.bai"),
"bigwigfiles": glob.glob("*/*.bw"),
}
publish_tracks(export_files)
will create a hub file at
:file:`<uploaddir>/OBFUSID/mapping/ucsc.hub`, where
OBFUSID is the obfuscated directory entry in the CGAT
download directory for a particular project.
If you want to create group tracks and get them to inherit from a
parent, you can supply an filename for a UCSC ini file. The ini
file defines two types of parameters, parents and set_features.
Parents define containers with a regex to identify the child
tracks. Set_features add additional features to all tracks
matching a regex. Parent and set_feature parameters are identified
by their respective "parent" or "set_features" prefixes.
For example, the following UCSC ini "test.ini" will create a
parent multiWig track called "Test" with the UCSC options as
defined in the values parameter. The values param must be a comma
separated list of key:value pairs which are seperated by a single
space. The regex param for parent_test defines the child tracks
which will be contained within "Test". The optional colour param
defines the colours for the child tracks. Colours are defined
using the brewer2mpl python module. Colour parameters must contain
the name of the pallete followed by the type of pallette.
The ini file below also defines a "set_features" parameter,
"bigwigs". Set_feature require a value and regex parameter. In
this case, the UCSC options in the values parameter will be added
to all tracks matching the ".*bigwig$" regex. As above, the values
param must be a comma separated list of key:value pairs which are
seperated by a single space. As above, an optional colours
parameter can also be given.
Note: colour palletes have a maximum number of allowable colours.
To see the available palletes and their size, run:
>import brewer2mpl
>brewer2mpl.print_maps()
>cat test.ini
#######################
#######################
[parent_test]
values=container multiWig,bigDataUrl Test,shortLabel Test,longLabel Test,type bigWig,viewLimits 0:160,visibility full,aggregate transparentOverlay,showSubtrackColorOnUi on,windowingFunction maximum,priority 1.2,configurable on,autoScale on,dragAndDrop subtracks
regex=.*-Saline-.*bw$
colour=Blues,Sequential
#######################
[set_features_bigwigs]
values=configurable on,autoScale on,useScore on,visibility full
regex=.*bigwig$
colour=Oranges,Sequential
#######################
#######################
Arguments
---------
export_files : dict
Dictionary mapping filetypes to files.
prefix : string
will be added to each track.
project_id : string
The project identifier. If not given, it will be taken from
the path of the project directory.
project_name : string
The project name, typically the project number. If not given,
it will be taken from the current directory.
'''
# the import is located here to avoid cyclical dependencies
# between Local.py, Pipeline.py and PipelineUCSC.py
import CGATPipelines.PipelineUCSC as PipelineUCSC
if not prefix:
prefix = PARAMS.get("report_prefix", "")
if not UCSC_ini:
UCSC_ini = PARAMS.get("ucsc_ini", None)
web_dir = PARAMS["web_dir"]
if project_id is None:
project_id = getProjectId()
if project_name is None:
project_name = getProjectName()
src_export = os.path.abspath("export")
dest_report = prefix + "report"
dest_export = prefix + "export"
hubdir = os.path.join(PARAMS["web_dir"], "ucsc")
if not os.path.exists(hubdir):
E.info("creating %s" % hubdir)
os.mkdir(hubdir)
# write the UCSC hub file
hubfile = os.path.join(hubdir, "hub.txt")
genomesfile = os.path.join(hubdir, "genomes.txt")
trackdir = os.path.join(hubdir, PARAMS["genome"])
trackfile = os.path.join(hubdir, PARAMS["genome"], "trackDb.txt")
trackrelpath = os.path.join(PARAMS["genome"], "trackDb.txt")
if os.path.exists(hubfile):
with IOTools.openFile(hubfile) as infile:
hubdata = PipelineUCSC.readUCSCFile(infile)
else:
hubdata = [('hub', "CGAT-" + project_name),
('shortLabel', "CGAT-" + project_name),
('longLabel', "Data for CGAT project %s" % project_name),
('genomesFile', "genomes.txt"),
('email', '*****@*****.**')]
E.info("writing to %s" % hubfile)
with IOTools.openFile(hubfile, "w") as outfile:
PipelineUCSC.writeUCSCFile(outfile, hubdata)
# create the genomes.txt file - append to it if necessary.
if os.path.exists(genomesfile):
with IOTools.openFile(genomesfile) as infile:
genomes = PipelineUCSC.readUCSCFile(infile)
else:
genomes = []
if ("genome", PARAMS["genome"]) not in genomes:
genomes.append(("genome", PARAMS["genome"]))
genomes.append(("trackDb", trackrelpath))
E.info("writing to %s" % genomesfile)
with IOTools.openFile(genomesfile, "w") as outfile:
PipelineUCSC.writeUCSCFile(outfile, genomes)
# create the track data
if not os.path.exists(trackdir):
os.mkdir(trackdir)
if os.path.exists(trackfile):
E.debug('reading existing tracks from %s' % trackfile)
with IOTools.openFile(trackfile) as infile:
tracks = PipelineUCSC.readTrackFile(infile)
else:
tracks = []
tracks = collections.OrderedDict(tracks)
def getName(name):
if name.endswith(".bam"):
return "bam", name
elif name.endswith(".bw") or name.endswith(".bigwig"):
return "bigWig", name
elif name.endswith(".bb") or name.endswith(".bigbed"):
return "bigBed", name
else:
return None, None
for targetdir, filenames in export_files.items():
for src in filenames:
dest = os.path.join(trackdir, prefix + os.path.basename(src))
dest = os.path.abspath(dest)
# create a symlink
if not os.path.exists(dest):
try:
os.symlink(os.path.abspath(src), dest)
except OSError, msg:
E.warn("could not create symlink from %s to %s: %s" %
(os.path.abspath(src), dest, msg))
ucsctype, trackname = getName(os.path.basename(dest))
# ignore invalid types and other files (.bai files, ...)
if ucsctype is None:
continue
tracks[trackname] = (("bigDataUrl", os.path.basename(dest)),
("shortLabel", trackname),
("longLabel", trackname),
("type", ucsctype))
else:
colours = None
for n, child in enumerate(children):
if make_group:
# make a parent and a copy of the child so we have
# two tracks, one grouped, one by itself
values = UCSC_PARAMS[param]
tracks[name] = [x.split(" ") for x in values.split(",")]
group_trackname = child + "_grouped"
tracks[group_trackname] = tracks[child]
tracks[group_trackname] += (("parent", name),)
else:
# just add the values to the child
values = UCSC_PARAMS[name + "_values"]
tracks[child] += tuple([x.split(" ") for x in values.split(",")])
if colours:
rgb = ",".join(map(str, colours[n]))
tracks[child] += (("color", rgb),)
if make_group:
tracks[group_trackname] += (("color", rgb),)
E.info("writing to %s" % trackfile)
with IOTools.openFile(trackfile, "w") as outfile:
PipelineUCSC.writeTrackFile(outfile, list(tracks.iteritems()))
E.info(
"data hub has been created at http://www.cgat.org/downloads/%(project_id)s/ucsc/hub.txt" % locals())
def publish_tracks(export_files,
prefix="",
project_id=None,
project_name=None):
'''publish a UCSC Track Hub.
*export_files* is a dictionary of filetypes and files.
*prefix* will be added to each track.
'''
if not prefix:
prefix = PARAMS.get("report_prefix", "")
web_dir = PARAMS["web_dir"]
if project_id is None:
project_id = P.getProjectId()
if project_name is None:
project_name = P.getProjectName()
src_export = os.path.abspath("export")
dest_report = prefix + "report"
dest_export = prefix + "export"
hubdir = os.path.join(PARAMS["web_dir"], "ucsc")
if not os.path.exists(hubdir):
E.info("creating %s" % hubdir)
os.mkdir(hubdir)
# write the UCSC hub file
hubfile = os.path.join(hubdir, "hub.txt")
genomesfile = os.path.join(hubdir, "genomes.txt")
trackdir = os.path.join(hubdir, PARAMS["genome"])
trackfile = os.path.join(hubdir, PARAMS["genome"], "trackDb.txt")
trackrelpath = os.path.join(PARAMS["genome"], "trackDb.txt")
if os.path.exists(hubfile):
with IOTools.openFile(hubfile) as infile:
hubdata = PipelineUCSC.readUCSCFile(infile)
else:
hubdata = [('hub', "CGAT-" + project_name),
('shortLabel', "CGAT-" + project_name),
('longLabel', "Data for CGAT project %s" % project_name),
('genomesFile', "genomes.txt"),
('email', '*****@*****.**')]
E.info("writing to %s" % hubfile)
with IOTools.openFile(hubfile, "w") as outfile:
PipelineUCSC.writeUCSCFile(outfile, hubdata)
# create the genomes.txt file - append to it if necessary.
if os.path.exists(genomesfile):
with IOTools.openFile(genomesfile) as infile:
genomes = PipelineUCSC.readUCSCFile(infile)
else:
genomes = []
if ("genome", PARAMS["genome"]) not in genomes:
genomes.append(("genome", PARAMS["genome"]))
genomes.append(("trackDb", trackrelpath))
E.info("writing to %s" % genomesfile)
with IOTools.openFile(genomesfile, "w") as outfile:
PipelineUCSC.writeUCSCFile(outfile, genomes)
# create the track data
if not os.path.exists(trackdir):
os.mkdir(trackdir)
if os.path.exists(trackfile):
E.debug('reading existing tracks from %s' % trackfile)
with IOTools.openFile(trackfile) as infile:
tracks = PipelineUCSC.readTrackFile(infile)
else:
tracks = []
tracks = collections.OrderedDict(tracks)
def getName(name):
if name.endswith(".bam"):
return "bam", name
elif name.endswith(".bw") or name.endswith(".bigwig"):
return "bigWig", name
else:
return None, None
for targetdir, filenames in export_files.items():
for src in filenames:
dest = os.path.join(trackdir, prefix + os.path.basename(src))
dest = os.path.abspath(dest)
# create a symlink
if not os.path.exists(dest):
try:
os.symlink(os.path.abspath(src), dest)
except OSError, msg:
E.warn("could not create symlink from %s to %s: %s" %
(os.path.abspath(src), dest, msg))
ucsctype, trackname = getName(os.path.basename(dest))
# ignore invalid types and other files (.bai files, ...)
if ucsctype is None:
continue
tracks[trackname] = (("bigDataUrl", os.path.basename(dest)),
("shortLabel", trackname),
("longLabel", trackname),
("type", ucsctype))
def publish_tracks(export_files, prefix="", project_id=None, project_name=None):
"""publish a UCSC Track Hub.
This method takes a dictionary of file types associated
with files. For each file, a link will be created in
the upload directory. The track will be stored under
a project name, which will be derived from the location
of the working directory.
Information about the genome, the upload directory, etc. will be
taken from the global configuration dictionary.
For example, calling the following code in a pipeline executed
in .../proj013/mapping::
export_files = {
"bamfiles": glob.glob("*/*.bam") + glob.glob("*/*.bam.bai"),
"bigwigfiles": glob.glob("*/*.bw"),
}
publish_tracks(export_files)
will create a hub file at
:file:`<uploaddir>/OBFUSID/mapping/ucsc.hub`, where
OBFUSID is the obfuscated directory entry in the CGAT
download directory for a particular project.
Arguments
---------
export_files : dict
Dictionary mapping filetypes to files.
prefix : string
will be added to each track.
project_id : string
The project identifier. If not given, it will be taken from
the path of the project directory.
project_name : string
The project name, typically the project number. If not given,
it will be taken from the current directory.
"""
# the import is located here to avoid cyclical dependencies
# between Local.py, Pipeline.py and PipelineUCSC.py
import CGATPipelines.PipelineUCSC as PipelineUCSC
if not prefix:
prefix = PARAMS.get("report_prefix", "")
web_dir = PARAMS["web_dir"]
if project_id is None:
project_id = getProjectId()
if project_name is None:
project_name = getProjectName()
src_export = os.path.abspath("export")
dest_report = prefix + "report"
dest_export = prefix + "export"
hubdir = os.path.join(PARAMS["web_dir"], "ucsc")
if not os.path.exists(hubdir):
E.info("creating %s" % hubdir)
os.mkdir(hubdir)
# write the UCSC hub file
hubfile = os.path.join(hubdir, "hub.txt")
genomesfile = os.path.join(hubdir, "genomes.txt")
trackdir = os.path.join(hubdir, PARAMS["genome"])
trackfile = os.path.join(hubdir, PARAMS["genome"], "trackDb.txt")
trackrelpath = os.path.join(PARAMS["genome"], "trackDb.txt")
if os.path.exists(hubfile):
with IOTools.openFile(hubfile) as infile:
hubdata = PipelineUCSC.readUCSCFile(infile)
else:
hubdata = [
("hub", "CGAT-" + project_name),
("shortLabel", "CGAT-" + project_name),
("longLabel", "Data for CGAT project %s" % project_name),
("genomesFile", "genomes.txt"),
("email", "*****@*****.**"),
]
E.info("writing to %s" % hubfile)
with IOTools.openFile(hubfile, "w") as outfile:
PipelineUCSC.writeUCSCFile(outfile, hubdata)
# create the genomes.txt file - append to it if necessary.
if os.path.exists(genomesfile):
with IOTools.openFile(genomesfile) as infile:
genomes = PipelineUCSC.readUCSCFile(infile)
else:
genomes = []
if ("genome", PARAMS["genome"]) not in genomes:
genomes.append(("genome", PARAMS["genome"]))
genomes.append(("trackDb", trackrelpath))
E.info("writing to %s" % genomesfile)
with IOTools.openFile(genomesfile, "w") as outfile:
PipelineUCSC.writeUCSCFile(outfile, genomes)
# create the track data
if not os.path.exists(trackdir):
os.mkdir(trackdir)
if os.path.exists(trackfile):
E.debug("reading existing tracks from %s" % trackfile)
with IOTools.openFile(trackfile) as infile:
tracks = PipelineUCSC.readTrackFile(infile)
else:
tracks = []
tracks = collections.OrderedDict(tracks)
def getName(name):
if name.endswith(".bam"):
return "bam", name
elif name.endswith(".bw") or name.endswith(".bigwig"):
return "bigWig", name
else:
return None, None
for targetdir, filenames in export_files.items():
for src in filenames:
dest = os.path.join(trackdir, prefix + os.path.basename(src))
dest = os.path.abspath(dest)
# create a symlink
if not os.path.exists(dest):
try:
os.symlink(os.path.abspath(src), dest)
except OSError, msg:
E.warn("could not create symlink from %s to %s: %s" % (os.path.abspath(src), dest, msg))
ucsctype, trackname = getName(os.path.basename(dest))
# ignore invalid types and other files (.bai files, ...)
if ucsctype is None:
continue
tracks[trackname] = (
("bigDataUrl", os.path.basename(dest)),
("shortLabel", trackname),
("longLabel", trackname),
("type", ucsctype),
)
def publish_tracks(export_files,
prefix="",
project_id=None,
project_name=None):
'''publish a UCSC Track Hub.
*export_files* is a dictionary of filetypes and files.
*prefix* will be added to each track.
'''
if not prefix:
prefix = PARAMS.get("report_prefix", "")
web_dir = PARAMS["web_dir"]
if project_id is None:
project_id = P.getProjectId()
if project_name is None:
project_name = P.getProjectName()
src_export = os.path.abspath("export")
dest_report = prefix + "report"
dest_export = prefix + "export"
hubdir = os.path.join(PARAMS["web_dir"], "ucsc")
if not os.path.exists(hubdir):
E.info("creating %s" % hubdir)
os.mkdir(hubdir)
# write the UCSC hub file
hubfile = os.path.join(hubdir, "hub.txt")
genomesfile = os.path.join(hubdir, "genomes.txt")
trackdir = os.path.join(hubdir, PARAMS["genome"])
trackfile = os.path.join(hubdir, PARAMS["genome"], "trackDb.txt")
trackrelpath = os.path.join(PARAMS["genome"], "trackDb.txt")
if os.path.exists(hubfile):
with IOTools.openFile(hubfile) as infile:
hubdata = PipelineUCSC.readUCSCFile(infile)
else:
hubdata = [('hub', "CGAT-" + project_name),
('shortLabel', "CGAT-" + project_name),
('longLabel', "Data for CGAT project %s" % project_name),
('genomesFile', "genomes.txt"),
('email', '*****@*****.**')]
E.info("writing to %s" % hubfile)
with IOTools.openFile(hubfile, "w") as outfile:
PipelineUCSC.writeUCSCFile(outfile, hubdata)
# create the genomes.txt file - append to it if necessary.
if os.path.exists(genomesfile):
with IOTools.openFile(genomesfile) as infile:
genomes = PipelineUCSC.readUCSCFile(infile)
else:
genomes = []
if ("genome", PARAMS["genome"]) not in genomes:
genomes.append(("genome", PARAMS["genome"]))
genomes.append(("trackDb", trackrelpath))
E.info("writing to %s" % genomesfile)
with IOTools.openFile(genomesfile, "w") as outfile:
PipelineUCSC.writeUCSCFile(outfile, genomes)
# create the track data
if not os.path.exists(trackdir):
os.mkdir(trackdir)
if os.path.exists(trackfile):
E.debug('reading existing tracks from %s' % trackfile)
with IOTools.openFile(trackfile) as infile:
tracks = PipelineUCSC.readTrackFile(infile)
else:
tracks = []
tracks = collections.OrderedDict(tracks)
def getName(name):
if name.endswith(".bam"):
return "bam", name
elif name.endswith(".bw") or name.endswith(".bigwig"):
return "bigWig", name
else:
return None, None
for targetdir, filenames in export_files.items():
for src in filenames:
dest = os.path.join(trackdir, prefix + os.path.basename(src))
dest = os.path.abspath(dest)
# create a symlink
if not os.path.exists(dest):
try:
os.symlink(os.path.abspath(src), dest)
except OSError, msg:
E.warn("could not create symlink from %s to %s: %s" %
(os.path.abspath(src), dest, msg))
ucsctype, trackname = getName(os.path.basename(dest))
# ignore invalid types and other files (.bai files, ...)
if ucsctype is None:
continue
tracks[trackname] = (("bigDataUrl", os.path.basename(dest)),
("shortLabel", trackname),
("longLabel", trackname),
("type", ucsctype))
return "bigWig", name
else:
return None, None
for targetdir, filenames in export_files.items():
for src in filenames:
dest = os.path.join(trackdir, prefix + os.path.basename(src))
dest = os.path.abspath(dest)
# create a symlink
if not os.path.exists(dest):
try:
os.symlink(os.path.abspath(src), dest)
except OSError, msg:
E.warn("could not create symlink from %s to %s: %s" %
(os.path.abspath(src), dest, msg))
ucsctype, trackname = getName(os.path.basename(dest))
# ignore invalid types and other files (.bai files, ...)
if ucsctype is None:
continue
tracks[trackname] = (("bigDataUrl", os.path.basename(dest)),
("shortLabel", trackname),
("longLabel", trackname),
("type", ucsctype))
E.info("writing to %s" % trackfile)
with IOTools.openFile(trackfile, "w") as outfile:
PipelineUCSC.writeTrackFile(outfile, list(tracks.iteritems()))
E.info(
"data hub has been created at http://www.cgat.org/downloads/%(project_id)s/ucsc/hub.txt" % locals())
def publish_tracks(export_files,
prefix="",
project_id=None,
project_name=None,
UCSC_ini=None):
'''publish a UCSC Track Hub.
This method takes a dictionary of file types associated
with files. For each file, a link will be created in
the upload directory. The track will be stored under
a project name, which will be derived from the location
of the working directory.
Information about the genome, the upload directory, etc. will be
taken from the global configuration dictionary.
For example, calling the following code in a pipeline executed
in .../proj013/mapping::
export_files = {
"bamfiles": glob.glob("*/*.bam") + glob.glob("*/*.bam.bai"),
"bigwigfiles": glob.glob("*/*.bw"),
}
publish_tracks(export_files)
will create a hub file at
:file:`<uploaddir>/OBFUSID/mapping/ucsc.hub`, where
OBFUSID is the obfuscated directory entry in the CGAT
download directory for a particular project.
If you want to create group tracks and get them to inherit from a
parent, you can supply an filename for a UCSC ini file. The ini
file defines two types of parameters, parents and set_features.
Parents define containers with a regex to identify the child
tracks. Set_features add additional features to all tracks
matching a regex. Parent and set_feature parameters are identified
by their respective "parent" or "set_features" prefixes.
For example, the following UCSC ini "test.ini" will create a
parent multiWig track called "Test" with the UCSC options as
defined in the values parameter. The values param must be a comma
separated list of key:value pairs which are seperated by a single
space. The regex param for parent_test defines the child tracks
which will be contained within "Test". The optional colour param
defines the colours for the child tracks. Colours are defined
using the brewer2mpl python module. Colour parameters must contain
the name of the pallete followed by the type of pallette.
The ini file below also defines a "set_features" parameter,
"bigwigs". Set_feature require a value and regex parameter. In
this case, the UCSC options in the values parameter will be added
to all tracks matching the ".*bigwig$" regex. As above, the values
param must be a comma separated list of key:value pairs which are
seperated by a single space. As above, an optional colours
parameter can also be given.
Note: colour palletes have a maximum number of allowable colours.
To see the available palletes and their size, run:
>import brewer2mpl
>brewer2mpl.print_maps()
>cat test.ini
#######################
#######################
[parent_test]
values=container multiWig,bigDataUrl Test,shortLabel Test,longLabel Test,type bigWig,viewLimits 0:160,visibility full,aggregate transparentOverlay,showSubtrackColorOnUi on,windowingFunction maximum,priority 1.2,configurable on,autoScale on,dragAndDrop subtracks
regex=.*-Saline-.*bw$
colour=Blues,Sequential
#######################
[set_features_bigwigs]
values=configurable on,autoScale on,useScore on,visibility full
regex=.*bigwig$
colour=Oranges,Sequential
#######################
#######################
Arguments
---------
export_files : dict
Dictionary mapping filetypes to files.
prefix : string
will be added to each track.
project_id : string
The project identifier. If not given, it will be taken from
the path of the project directory.
project_name : string
The project name, typically the project number. If not given,
it will be taken from the current directory.
'''
# the import is located here to avoid cyclical dependencies
# between Local.py, Pipeline.py and PipelineUCSC.py
import CGATPipelines.PipelineUCSC as PipelineUCSC
if not prefix:
prefix = PARAMS.get("report_prefix", "")
if not UCSC_ini:
UCSC_ini = PARAMS.get("ucsc_ini", None)
web_dir = PARAMS["web_dir"]
if project_id is None:
project_id = getProjectId()
if project_name is None:
project_name = getProjectName()
src_export = os.path.abspath("export")
dest_report = prefix + "report"
dest_export = prefix + "export"
hubdir = os.path.join(PARAMS["web_dir"], "ucsc")
if not os.path.exists(hubdir):
E.info("creating %s" % hubdir)
os.mkdir(hubdir)
# write the UCSC hub file
hubfile = os.path.join(hubdir, "hub.txt")
genomesfile = os.path.join(hubdir, "genomes.txt")
trackdir = os.path.join(hubdir, PARAMS["genome"])
trackfile = os.path.join(hubdir, PARAMS["genome"], "trackDb.txt")
trackrelpath = os.path.join(PARAMS["genome"], "trackDb.txt")
if os.path.exists(hubfile):
with IOTools.openFile(hubfile) as infile:
hubdata = PipelineUCSC.readUCSCFile(infile)
else:
hubdata = [('hub', "CGAT-" + project_name),
('shortLabel', "CGAT-" + project_name),
('longLabel', "Data for CGAT project %s" % project_name),
('genomesFile', "genomes.txt"),
('email', '*****@*****.**')]
E.info("writing to %s" % hubfile)
with IOTools.openFile(hubfile, "w") as outfile:
PipelineUCSC.writeUCSCFile(outfile, hubdata)
# create the genomes.txt file - append to it if necessary.
if os.path.exists(genomesfile):
with IOTools.openFile(genomesfile) as infile:
genomes = PipelineUCSC.readUCSCFile(infile)
else:
genomes = []
if ("genome", PARAMS["genome"]) not in genomes:
genomes.append(("genome", PARAMS["genome"]))
genomes.append(("trackDb", trackrelpath))
E.info("writing to %s" % genomesfile)
with IOTools.openFile(genomesfile, "w") as outfile:
PipelineUCSC.writeUCSCFile(outfile, genomes)
# create the track data
if not os.path.exists(trackdir):
os.mkdir(trackdir)
if os.path.exists(trackfile):
E.debug('reading existing tracks from %s' % trackfile)
with IOTools.openFile(trackfile) as infile:
tracks = PipelineUCSC.readTrackFile(infile)
else:
tracks = []
tracks = collections.OrderedDict(tracks)
def getName(name):
if name.endswith(".bam"):
return "bam", name
elif name.endswith(".bw") or name.endswith(".bigwig"):
return "bigWig", name
elif name.endswith(".bb") or name.endswith(".bigbed"):
return "bigBed", name
else:
return None, None
for targetdir, filenames in export_files.items():
for src in filenames:
dest = os.path.join(trackdir, prefix + os.path.basename(src))
dest = os.path.abspath(dest)
# create a symlink
if not os.path.exists(dest):
try:
os.symlink(os.path.abspath(src), dest)
except OSError, msg:
E.warn("could not create symlink from %s to %s: %s" %
(os.path.abspath(src), dest, msg))
ucsctype, trackname = getName(os.path.basename(dest))
# ignore invalid types and other files (.bai files, ...)
if ucsctype is None:
continue
tracks[trackname] = (("bigDataUrl", os.path.basename(dest)),
("shortLabel", trackname),
("longLabel", trackname), ("type", ucsctype))
