def transcribe(input_file_url):
"""
Transcribes an audio file. The audio file format is a URL.
Params:
input_file_url (string): The input audio file URL
"""
# Setting up environment
if not utilities.check_env_vars():
return
utilities.create_ancillary_folders()
# download the podcast file
filepath = get_podcast_file(input_file_url)
# convert file to raw audio chunks
chunks = convert_to_raw_audio_chunks(filepath)
# transcribe chunks
transcriber = Transcriber(os.environ['GOOGLE_API_KEY'])
transcript = transcriber.transcribe_many(chunks)
# write to the output file
output_file_name = os.path.split(filepath)[-1]
utilities.write_output_file(output_file_name, transcript)
print "Cleaning up...\n"
utilities.cleanup()
def add_to_db(Session, k):
session = Session()
t = dt.date.today()
adds = 0
cleantracks = k
for line in cleantracks:
clean_name = cleanup(line[0])
n_ = band(name=line[0],
song=line[1],
album = line[2],
release_year = line[3],
source=line[4],
dateplayed=line[5],
dateadded=t,
cleanname=clean_name)
q = session.query(band).filter(band.name == n_.name, band.song == n_.song, band.source == n_.source)
if q.first() == None:
session.add(n_)
adds += 1
else:
try:
print ('Already had {0} - {1}'.format(n_.name, n_.song))
except:
print ('Already had it. Cannot print. ID is {0}'.format(q.first().id))
session.commit()
return adds
def load_to_db(albumlist):
socket.setdefaulttimeout(15)
# creation of the SQL database and the "session" object that is used to manage
# communications with the database
engine = create_engine('sqlite:///../../databases/scout.db')
session_factory = sessionmaker(bind=engine)
Session = scoped_session(session_factory)
metadata = MetaData(db)
db.metadata.create_all(engine)
session = Session()
t = dt.date.today()
adds = 0
for i in albumlist:
print (i)
clean_name = cleanup(i[0])
n_ = band(name=i[0],
album=i[1],
source='KEXP Countdown 2018',
appeared ='KEXP Countdown 2018',
dateadded=t,
cleanname=clean_name)
q = session.query(band).filter(band.name == n_.name, band.song == n_.song)
if q.first() == None:
session.add(n_)
adds += 1
else:
try:
print ('Already had {0} - {1}'.format(n_.name, n_.song))
except:
print ('Already had it. Cannot print. ID is {0}'.format(q.first().id))
session.commit()
print ('Added {0} songs'.format(adds))
def load_other_bands(Session, choices):
# this loop pulls down band names from the sources identified.
# each source, however, needs its own function, since the
# websites are set up differently
proceed = False
for src in choices:
if src in bandsources:
# at least one of the choices needs to be in bandsources
# or else this function doesn't need to proceed
proceed = True
if proceed == False:
return
session = Session()
today = dt.date.today()
other_sources = [
'KEXP Music That Matters', 'Pitchfork Top Tracks', 'Stereogum',
'Metacritic', 'KCRW', 'Pitchfork', 'KEXP charts', 'KNKX'
]
other_choices = [x for x in choices if x in other_sources]
for src in other_choices:
print('\n\n\n', src)
try:
list = grabbands(src)
except Exception as e:
print('{0} not provided for in load_other_bands.grabbands'.format(
src))
print(str(e))
add_count = 0
for i in list:
h = cleanup(i.name)
if h != '':
try:
q = session.query(band).filter(band.cleanname == h)
if q.first() == None:
i.source = src
i.cleanname = h
i.dateadded = today
print("Adding {0} (from {1})".format(i.name, i.source))
session.add(i)
session.commit()
add_count += 1
except Exception as e:
print(str(e))
print('Added {0} entries to the {1} collection. \n\n'.format(
add_count, src))
return
msg['content']['url'])
for category in msg['categories']:
if category not in pre_payload:
pre_payload[category] = []
pre_payload[category].append(msg)
payload = [dblock()]
for category in pre_payload:
payload.append(cblock(category))
pre_payload[category].sort(key=lambda x: x['type'])
types = set([t['type'] for t in pre_payload[category]])
for t in types:
payload.append(tblock(t))
payload.append(
iblock(
[msg for msg in pre_payload[category] if msg['type'] == t],
category, t))
payload.append(dblock())
ec.parse()
slacky.send_message(blocks=payload, channel_id=channel_id)
slacky.delete_set_messages(messages, channel_id=channel_id)
ut.cleanup()
def similar(a, b):
a = cleanup(a)
b = cleanup(b)
return SequenceMatcher(None, a, b).ratio()
def main(argv=sys.argv[1:]):
parser = OptionParser(usage=USAGE, version="Version: " + __version__)
# Required arguments
requiredOptions = OptionGroup(
parser, "Required options",
"These options are required to run BinGeR, and may be supplied in any order."
)
requiredOptions.add_option(
"-l",
"--sample_list",
type="string",
metavar="FILE",
help="Text file containing all sample names, one per line")
requiredOptions.add_option(
"-o",
"--out_dir",
type="string",
metavar="OUTDIR",
help=
"Working directory where the results and intermediate files will be stored at"
)
parser.add_option_group(requiredOptions)
# Optional arguments that need to be supplied if not the same as default
optOptions = OptionGroup(
parser, "Optional parameters",
"There options are optional, and may be supplied in any order.")
optOptions.add_option(
"-b",
"--bams_dir",
type="string",
default="Bams",
metavar="DIR",
help=
"Directory where sorted bam files (reads versus assembly, same sample) are, the naming should follow \"sample.*.bam\" convention. [Default: ./Bams]"
)
optOptions.add_option(
"-c",
"--coverage_dir",
type="string",
default="Coverage",
metavar="DIR",
help=
"Directory where coverage files are, naming follows \"sampleA.vs.sampleB.*.coverage\" convention. [Default: ./Coverage]"
)
optOptions.add_option(
"-a",
"--assemblies_dir",
type="string",
default="Assemblies",
metavar="DIR",
help=
"Directory where assemblies in fasta format are, naming follows \"sample.*.fa\" convention. [Default: ./Assemblies]"
)
optOptions.add_option(
"-z",
"--zscore_dir",
type="string",
default="ZScores",
metavar="DIR",
help=
"Directory where oligo-nt z-score files are, naming follows \"sample.*.ZScore\" convention. [Default: ./ZScore]"
)
optOptions.add_option(
"-s",
"--hmmscan_dir",
type="string",
default="HMMScan",
metavar="DIR",
help=
"Directory where hmmscan files are, naming follows \"sample.*.hmmscan\" convention. [Default: ./HMMScan]"
)
optOptions.add_option(
"-t",
"--num_proc",
type="int",
default=1,
metavar='INT',
help="Number of processor for BinGeR to use [default: 1].")
optOptions.add_option("--blat",
type="string",
default="blat",
help="Path to blat, specify if not in env.")
parser.add_option_group(optOptions)
# Binning parameters that could fine tune the process
clusteringOptions = OptionGroup(
parser, "Binning parameters",
"There options are optional, and may be supplied in any order.")
clusteringOptions.add_option(
"-m",
"--min_core",
type="int",
default=1e5,
metavar='INT',
help="Minimum size to consider as bin core [default: 1e5].")
clusteringOptions.add_option(
"-u",
"--cov_clustering_min_length",
dest="minCovLength",
type="int",
default=1500,
metavar='INT',
help=
"Minimum contig length to be considered in coverage clustering [default: 1500]."
)
clusteringOptions.add_option(
"--min_cov_corrcoef",
dest="minCovCorrceof",
type="float",
default=0.95,
metavar='FLOAT',
help=
"Minimum correlation coefficient cutoff for form a link between contigs using coverage profiles [default: 0.95]."
)
clusteringOptions.add_option(
"--min_zscore_corrcoef",
dest="minZScoreCorrceof",
type="float",
default=0.95,
metavar='FLOAT',
help=
"Minimum correlation coefficient cutoff for form a link between contigs using tri-/tetra-nt frequency Z-Score [default: 0.90]."
)
clusteringOptions.add_option(
"-x",
"--zscore_clustering_min_length",
dest="minZLength",
type="int",
default=3000,
metavar='INT',
help=
"Minimum contig length to be considered in Z-score clustering [default: 2000]."
)
clusteringOptions.add_option(
"-d",
"--cpr_alpha",
type="float",
default=0.9,
metavar='FLOAT',
help=
"The dampening factor, alpha, in community personalized PageRank [default: 0.9, range: (0.75, 0.95)]."
)
clusteringOptions.add_option(
"-e",
"--cpr_tol",
type="float",
default=1e-4,
metavar='FLOAT',
help=
"The error tolerance factor, tol, in community personalized PageRank [default: 1e-4, range: (1e-8, 1e-2)]."
)
clusteringOptions.add_option(
"-i",
"--cpr_maxiter",
type="int",
default=50,
metavar='INT',
help=
"The max iterations performed in community personalized PageRank [default: 50, range: (20, 100)]."
)
parser.add_option_group(clusteringOptions)
# runtime settings that could affect the file saving and message printing
runtimeSettings = OptionGroup(
parser, "Runtime settings",
"There options are optional, and may be supplied in any order.")
runtimeSettings.add_option(
"-q",
"--quiet",
default=False,
action="store_true",
help=
"Suppress printing detailed runtime information, only important messages will show [default: False]."
)
runtimeSettings.add_option(
"--no_intermediates",
action="store_false",
dest="save_intermediates",
default=True,
help=
"Do no save intermediate files during runtime [default: True (save intermediates)]."
)
parser.add_option_group(runtimeSettings)
(options, args) = parser.parse_args(argv)
if options.sample_list is None:
parser.error("A list of samples and a working directory are required!")
exit(0)
if options.out_dir is None:
parser.error("An output directory is required to supply!")
exit(0)
if options.num_proc < 1:
parser.error(
"Number of processors must be integer >= 1, you supplied %i" %
options.num_proc)
exit(0)
if options.min_core < 1e4 or options.min_core > 1e6:
parser.error(
"Size of minimum bin size must be in range [1e4, 1e6]bp, you supplied %i"
% options.min_core)
exit(0)
if options.cpr_alpha < 0.75 or options.cpr_alpha > 0.95:
parser.error(
"Community PageRank Alpha must be a float in range [0.75, 0.95], you supplied %.3f"
% options.cpr_alpha)
exit(0)
if options.cpr_tol < 1e-8 or options.cpr_tol > 1e-2:
parser.error(
"Community PageRank tol must be a float in range [1e-8, 1e-2], you supplied %.3f"
% options.cpr_tol)
exit(0)
if options.cpr_maxiter < 20 or options.cpr_maxiter > 100:
parser.error(
"Community PageRank tol must be a float in range [20, 100], you supplied %i"
% options.cpr_maxiter)
exit(0)
total_start_time = time()
sys.stdout.write("BinGeR started at %s\n" % (ctime()))
sys.stdout.flush()
# test if blat exists
blatTest = Popen(options.blat, shell=True, stdout=PIPE).stdout.read()
if blatTest == None or len(blatTest) == 0:
sys.stderr.write("FATAL: blat not found in path!")
exit(0)
# check sanity of the files in required directories
projInfo = ProjectInfo()
projInfo.initProject(options)
if not options.quiet:
projInfo.printSamples()
# build networkx graph for Project
sys.stdout.write('\nInitializing contig space...\n')
G = cSpace.ContigSpace(projInfo.samples)
initCoresPath = projInfo.out_dir + '/initCores'
if os.path.exists(initCoresPath):
if len(glob.glob(initCoresPath + '/*.cpickle')) > 0:
G.refineCores(projInfo, options)
G.recruitContigs(projInfo, options)
else:
G.initSubgraphs(projInfo, options)
G.forgeCores(projInfo, options)
G.refineCores(projInfo, options)
G.recruitContigs(projInfo, options)
# output bins and the evaluation, extract reads of bins for downstream analysis.
utilities.outputBins(projInfo, options)
# get all the reads for the bins
utilities.extractReadsForBins(projInfo, options)
# clean up if necessary
if not options.save_intermediates:
utilities.cleanup(projInfo)
total_finish_time = time()
sys.stdout.write("BinGeR finished at %s\n" % (ctime()))
sys.stdout.flush()
return
def main(argv = sys.argv[1:]):
parser = OptionParser(usage = USAGE, version="Version: " + __version__)
# Required arguments
requiredOptions = OptionGroup(parser, "Required options",
"These options are required to run BinGeR, and may be supplied in any order.")
requiredOptions.add_option("-l", "--sample_list", type = "string", metavar = "FILE",
help = "Text file containing all sample names, one per line")
requiredOptions.add_option("-o", "--out_dir", type = "string", metavar = "OUTDIR",
help = "Working directory where the results and intermediate files will be stored at")
parser.add_option_group(requiredOptions)
# Optional arguments that need to be supplied if not the same as default
optOptions = OptionGroup(parser, "Optional parameters",
"There options are optional, and may be supplied in any order.")
optOptions.add_option("-b", "--bams_dir", type = "string", default = "Bams", metavar = "DIR",
help = "Directory where sorted bam files (reads versus assembly, same sample) are, the naming should follow \"sample.*.bam\" convention. [Default: ./Bams]")
optOptions.add_option("-c", "--coverage_dir", type = "string", default = "Coverage", metavar = "DIR",
help = "Directory where coverage files are, naming follows \"sampleA.vs.sampleB.*.coverage\" convention. [Default: ./Coverage]")
optOptions.add_option("-a", "--assemblies_dir", type = "string", default = "Assemblies", metavar = "DIR",
help = "Directory where assemblies in fasta format are, naming follows \"sample.*.fa\" convention. [Default: ./Assemblies]")
optOptions.add_option("-z", "--zscore_dir", type = "string", default = "ZScores", metavar = "DIR",
help = "Directory where oligo-nt z-score files are, naming follows \"sample.*.ZScore\" convention. [Default: ./ZScore]")
optOptions.add_option("-s", "--hmmscan_dir", type = "string", default = "HMMScan", metavar = "DIR",
help = "Directory where hmmscan files are, naming follows \"sample.*.hmmscan\" convention. [Default: ./HMMScan]")
optOptions.add_option("-t", "--num_proc", type = "int", default = 1, metavar = 'INT',
help = "Number of processor for BinGeR to use [default: 1].")
optOptions.add_option("--blat", type = "string", default = "blat",
help = "Path to blat, specify if not in env.")
parser.add_option_group(optOptions)
# Binning parameters that could fine tune the process
clusteringOptions = OptionGroup(parser, "Binning parameters",
"There options are optional, and may be supplied in any order.")
clusteringOptions.add_option("-m", "--min_core", type = "int", default = 1e5, metavar = 'INT',
help = "Minimum size to consider as bin core [default: 1e5].")
clusteringOptions.add_option("-u", "--cov_clustering_min_length", dest = "minCovLength",
type = "int", default = 1500, metavar = 'INT',
help = "Minimum contig length to be considered in coverage clustering [default: 1500].")
clusteringOptions.add_option("--min_cov_corrcoef", dest = "minCovCorrceof",
type = "float", default = 0.95, metavar = 'FLOAT',
help = "Minimum correlation coefficient cutoff for form a link between contigs using coverage profiles [default: 0.95].")
clusteringOptions.add_option("--min_zscore_corrcoef", dest = "minZScoreCorrceof",
type = "float", default = 0.95, metavar = 'FLOAT',
help = "Minimum correlation coefficient cutoff for form a link between contigs using tri-/tetra-nt frequency Z-Score [default: 0.90].")
clusteringOptions.add_option("-x", "--zscore_clustering_min_length", dest = "minZLength",
type = "int", default = 3000, metavar = 'INT',
help = "Minimum contig length to be considered in Z-score clustering [default: 2000].")
clusteringOptions.add_option("-d", "--cpr_alpha", type = "float", default = 0.9, metavar = 'FLOAT',
help = "The dampening factor, alpha, in community personalized PageRank [default: 0.9, range: (0.75, 0.95)].")
clusteringOptions.add_option("-e", "--cpr_tol", type = "float", default = 1e-4, metavar = 'FLOAT',
help = "The error tolerance factor, tol, in community personalized PageRank [default: 1e-4, range: (1e-8, 1e-2)].")
clusteringOptions.add_option("-i", "--cpr_maxiter", type = "int", default = 50, metavar = 'INT',
help = "The max iterations performed in community personalized PageRank [default: 50, range: (20, 100)].")
parser.add_option_group(clusteringOptions)
# runtime settings that could affect the file saving and message printing
runtimeSettings = OptionGroup(parser, "Runtime settings",
"There options are optional, and may be supplied in any order.")
runtimeSettings.add_option("-q", "--quiet", default = False, action = "store_true",
help = "Suppress printing detailed runtime information, only important messages will show [default: False].")
runtimeSettings.add_option("--no_intermediates", action="store_false",
dest = "save_intermediates", default = True,
help = "Do no save intermediate files during runtime [default: True (save intermediates)].")
parser.add_option_group(runtimeSettings)
(options, args) = parser.parse_args(argv)
if options.sample_list is None:
parser.error("A list of samples and a working directory are required!")
exit(0)
if options.out_dir is None:
parser.error("An output directory is required to supply!")
exit(0)
if options.num_proc < 1:
parser.error("Number of processors must be integer >= 1, you supplied %i" % options.num_proc)
exit(0)
if options.min_core < 1e4 or options.min_core > 1e6:
parser.error("Size of minimum bin size must be in range [1e4, 1e6]bp, you supplied %i" % options.min_core)
exit(0)
if options.cpr_alpha < 0.75 or options.cpr_alpha > 0.95:
parser.error("Community PageRank Alpha must be a float in range [0.75, 0.95], you supplied %.3f" % options.cpr_alpha)
exit(0)
if options.cpr_tol < 1e-8 or options.cpr_tol > 1e-2:
parser.error("Community PageRank tol must be a float in range [1e-8, 1e-2], you supplied %.3f" % options.cpr_tol)
exit(0)
if options.cpr_maxiter < 20 or options.cpr_maxiter > 100:
parser.error("Community PageRank tol must be a float in range [20, 100], you supplied %i" % options.cpr_maxiter)
exit(0)
total_start_time = time()
sys.stdout.write("BinGeR started at %s\n"%(ctime()))
sys.stdout.flush()
# test if blat exists
blatTest = Popen(options.blat, shell=True, stdout=PIPE).stdout.read()
if blatTest == None or len(blatTest) == 0:
sys.stderr.write("FATAL: blat not found in path!")
exit(0)
# check sanity of the files in required directories
projInfo = ProjectInfo()
projInfo.initProject(options)
if not options.quiet:
projInfo.printSamples()
# build networkx graph for Project
sys.stdout.write('\nInitializing contig space...\n')
G = cSpace.ContigSpace(projInfo.samples)
initCoresPath = projInfo.out_dir + '/initCores'
if os.path.exists(initCoresPath):
if len(glob.glob(initCoresPath + '/*.cpickle')) > 0:
G.refineCores(projInfo, options)
G.recruitContigs(projInfo, options)
else:
G.initSubgraphs(projInfo, options)
G.forgeCores(projInfo, options)
G.refineCores(projInfo, options)
G.recruitContigs(projInfo, options)
# output bins and the evaluation, extract reads of bins for downstream analysis.
utilities.outputBins(projInfo, options)
# get all the reads for the bins
utilities.extractReadsForBins(projInfo, options)
# clean up if necessary
if not options.save_intermediates:
utilities.cleanup(projInfo)
total_finish_time = time()
sys.stdout.write("BinGeR finished at %s\n"%(ctime()))
sys.stdout.flush()
return
