def check_options(opts):
if not opts.workdir: raise Exception('ERROR: output option required.')
# check resume
if not path.exists(opts.workdir) and opts.resume:
print('WARNING: can use output files, found, not resuming...')
opts.resume = False
# sort filters
if opts.apply:
opts.apply.sort()
# for lustre file system....
if 'tmpdb' in opts and opts.tmpdb:
dbdir = opts.tmpdb
# tmp file
dbfile = 'trace_%s' % (''.join(
[ascii_letters[int(random() * 52)] for _ in range(10)]))
opts.tmpdb = path.join(dbdir, dbfile)
try:
copyfile(path.join(opts.workdir, 'trace.db'), opts.tmpdb)
except IOError:
pass
# check if job already run using md5 digestion of parameters
if already_run(opts) and not opts.force:
if 'tmpdb' in opts and opts.tmpdb:
remove(path.join(dbdir, dbfile))
exit('WARNING: exact same job already computed, see JOBs table above')
def check_options(opts):
if not opts.workdir: raise Exception('ERROR: output option required.')
# check resume
if not path.exists(opts.workdir) and opts.resume:
print ('WARNING: can use output files, found, not resuming...')
opts.resume = False
# sort filters
if opts.apply:
opts.apply.sort()
# for lustre file system....
if 'tmpdb' in opts and opts.tmpdb:
dbdir = opts.tmpdb
# tmp file
dbfile = 'trace_%s' % (''.join([ascii_letters[int(random() * 52)]
for _ in range(10)]))
opts.tmpdb = path.join(dbdir, dbfile)
try:
copyfile(path.join(opts.workdir, 'trace.db'), opts.tmpdb)
except IOError:
pass
# check if job already run using md5 digestion of parameters
if already_run(opts) and not opts.force:
if 'tmpdb' in opts and opts.tmpdb:
remove(path.join(dbdir, dbfile))
exit('WARNING: exact same job already computed, see JOBs table above')
def check_options(opts):
mkdir(opts.workdir)
# transform filtering reads option
opts.filter = filters_to_bin(opts.filter)
# enlighten plotting parameter writing
if opts.only_plot:
opts.plot = True
if opts.interactive:
if opts.nox:
raise Exception('ERROR: no screen no fun.\n'
'Interactive plot incompatible with noX option.')
opts.plot = True
opts.only_plot = True
# check resume
if not path.exists(opts.workdir):
raise IOError('ERROR: workdir not found.')
# check resume
if opts.triangular and opts.coord2:
raise NotImplementedError('ERROR: triangular is only available for '
'symmetric matrices.')
# for LUSTRE file system....
if 'tmpdb' in opts and opts.tmpdb:
dbdir = opts.tmpdb
# tmp file
dbfile = 'trace_%s' % (''.join(
[ascii_letters[int(random() * 52)] for _ in range(10)]))
opts.tmpdb = path.join(dbdir, dbfile)
try:
copyfile(path.join(opts.workdir, 'trace.db'), opts.tmpdb)
except IOError:
pass
# number of cpus
if opts.cpus == 0:
opts.cpus = cpu_count()
else:
opts.cpus = min(opts.cpus, cpu_count())
# check if job already run using md5 digestion of parameters
try:
if already_run(opts):
if not opts.force:
if 'tmpdb' in opts and opts.tmpdb:
remove(path.join(dbdir, dbfile))
exit(
'WARNING: exact same job already computed, see JOBs table above'
)
else:
warn(
'WARNING: exact same job already computed, overwriting...')
except IOError:
warn((""
"\nWARNING:\n new working directory created. It's ok... "
"but next time use TADbit since the beginning!! :)"))
def check_options(opts):
# check resume
if not path.exists(opts.workdir):
raise IOError('ERROR: wordir not found.')
if already_run(opts) and not opts.force:
exit('WARNING: exact same job already computed, see JOBs table above')
def check_options(opts):
mkdir(opts.workdir)
# transform filtering reads option
opts.filter = filters_to_bin(opts.filter)
# enlighten plotting parameter writing
if opts.only_plot:
opts.plot = True
if opts.interactive:
if opts.nox:
raise Exception('ERROR: no screen no fun.\n'
'Interactive plot incompatible with noX option.')
opts.plot = True
opts.only_plot = True
# check resume
if not path.exists(opts.workdir):
raise IOError('ERROR: workdir not found.')
# check resume
if opts.triangular and opts.coord2:
raise NotImplementedError('ERROR: triangular is only available for '
'symmetric matrices.')
# for LUSTRE file system....
if 'tmpdb' in opts and opts.tmpdb:
dbdir = opts.tmpdb
# tmp file
dbfile = 'trace_%s' % (''.join([ascii_letters[int(random() * 52)]
for _ in range(10)]))
opts.tmpdb = path.join(dbdir, dbfile)
try:
copyfile(path.join(opts.workdir, 'trace.db'), opts.tmpdb)
except IOError:
pass
# number of cpus
if opts.cpus == 0:
opts.cpus = cpu_count()
else:
opts.cpus = min(opts.cpus, cpu_count())
# check if job already run using md5 digestion of parameters
try:
if already_run(opts):
if not opts.force:
if 'tmpdb' in opts and opts.tmpdb:
remove(path.join(dbdir, dbfile))
exit('WARNING: exact same job already computed, see JOBs table above')
else:
warn('WARNING: exact same job already computed, overwriting...')
except IOError:
warn((""
"\nWARNING:\n new working directory created. It's ok... "
"but next time use TADbit from the beginning!! :)"))
def check_options(opts):
mkdir(opts.workdir)
# create empty DB if don't exists
dbpath = path.join(opts.workdir, 'trace.db')
open(dbpath, 'a').close()
# for lustre file system....
if 'tmpdb' in opts and opts.tmpdb:
dbdir = opts.tmpdb
# tmp file
dbfile = 'trace_%s' % (''.join([ascii_letters[int(random() * 52)]
for _ in range(10)]))
opts.tmpdb = path.join(dbdir, dbfile)
try:
copyfile(path.join(opts.workdir, 'trace.db'), opts.tmpdb)
except IOError:
pass
if opts.workdir1:
# tmp file
dbfile1 = 'trace1_%s' % (''.join([ascii_letters[int(random() * 52)]
for _ in range(10)]))
opts.tmpdb1 = path.join(dbdir, dbfile1)
try:
copyfile(path.join(opts.workdir1, 'trace.db'), opts.tmpdb1)
except IOError:
pass
if opts.workdir2:
# tmp file
dbfile2 = 'trace2_%s' % (''.join([ascii_letters[int(random() * 52)]
for _ in range(10)]))
opts.tmpdb2 = path.join(dbdir, dbfile2)
try:
copyfile(path.join(opts.workdir2, 'trace.db'), opts.tmpdb2)
except IOError:
pass
else:
if opts.workdir1:
opts.tmpdb1 = path.join(opts.workdir1, 'trace.db')
if opts.workdir2:
opts.tmpdb2 = path.join(opts.workdir2, 'trace.db')
# resolution needed to compare
if not opts.skip_comparison and not opts.reso:
raise Exception('ERROR: need to define resolution at which to compare')
# check if job already run using md5 digestion of parameters
if already_run(opts):
if 'tmpdb' in opts and opts.tmpdb:
remove(path.join(dbdir, dbfile))
if opts.workdir1:
remove(path.join(dbdir, dbfile1))
if opts.workdir2:
remove(path.join(dbdir, dbfile2))
exit('WARNING: exact same job already computed, see JOBs table above')
def check_options(opts):
mkdir(opts.workdir)
# create empty DB if don't exists
dbpath = path.join(opts.workdir, 'trace.db')
open(dbpath, 'a').close()
# for lustre file system....
if 'tmpdb' in opts and opts.tmpdb:
dbdir = opts.tmpdb
# tmp file
dbfile = 'trace_%s' % (''.join([ascii_letters[int(random() * 52)]
for _ in range(10)]))
opts.tmpdb = path.join(dbdir, dbfile)
try:
copyfile(path.join(opts.workdir, 'trace.db'), opts.tmpdb)
except IOError:
pass
if opts.workdir1:
# tmp file
dbfile1 = 'trace1_%s' % (''.join([ascii_letters[int(random() * 52)]
for _ in range(10)]))
opts.tmpdb1 = path.join(dbdir, dbfile1)
try:
copyfile(path.join(opts.workdir1, 'trace.db'), opts.tmpdb1)
except IOError:
pass
if opts.workdir2:
# tmp file
dbfile2 = 'trace2_%s' % (''.join([ascii_letters[int(random() * 52)]
for _ in range(10)]))
opts.tmpdb2 = path.join(dbdir, dbfile2)
try:
copyfile(path.join(opts.workdir2, 'trace.db'), opts.tmpdb2)
except IOError:
pass
else:
if opts.workdir1:
opts.tmpdb1 = path.join(opts.workdir1, 'trace.db')
if opts.workdir2:
opts.tmpdb2 = path.join(opts.workdir2, 'trace.db')
# resolution needed to compare
if not opts.skip_comparison and not opts.reso:
raise Exception('ERROR: need to define resolution at which to compare')
# check if job already run using md5 digestion of parameters
if already_run(opts):
if 'tmpdb' in opts and opts.tmpdb:
remove(path.join(dbdir, dbfile))
if opts.workdir1:
remove(path.join(dbdir, dbfile1))
if opts.workdir2:
remove(path.join(dbdir, dbfile2))
exit('WARNING: exact same job already computed, see JOBs table above')
def check_options(opts):
if not opts.workdir: raise Exception('ERROR: output option required.')
if opts.type != 'map':
raise NotImplementedError('ERROR: not yet there')
if not opts.genome: raise Exception('ERROR: genome parameter required.')
if not opts.workdir: raise Exception('ERROR: workdir parameter required.')
# check skip
if not path.exists(opts.workdir) and opts.skip:
print ('WARNING: can use output files, found, not skipping...')
opts.skip = False
if opts.workdir.endswith('/'):
opts.workdir = opts.workdir[:-1]
# write log
log_format = '[PARSING] %(message)s'
# reset logging
logging.getLogger().handlers = []
try:
print 'Writting log to ' + path.join(opts.workdir, 'process.log')
logging.basicConfig(level=logging.INFO,
format=log_format,
filename=path.join(opts.workdir, 'process.log'),
filemode='aw')
except IOError:
logging.basicConfig(level=logging.DEBUG,
format=log_format,
filename=path.join(opts.workdir, 'process.log2'),
filemode='aw')
# to display log on stdout also
logging.getLogger().addHandler(logging.StreamHandler())
# write version log
vlog_path = path.join(opts.workdir, 'TADbit_and_dependencies_versions.log')
dependencies = get_dependencies_version()
if not path.exists(vlog_path) or open(vlog_path).readlines() != dependencies:
logging.info('Writting versions of TADbit and dependencies')
vlog = open(vlog_path, 'w')
vlog.write(dependencies)
vlog.close()
# check if job already run using md5 digestion of parameters
if already_run(opts):
exit('WARNING: exact same job already computed, see JOBs table above')
def check_options(opts):
if not opts.workdir: raise Exception('ERROR: output option required.')
# check resume
if not path.exists(opts.workdir) and opts.resume:
print ('WARNING: can use output files, found, not resuming...')
opts.resume = False
# sort filters
if opts.apply:
opts.apply.sort()
# check if job already run using md5 digestion of parameters
if already_run(opts) and not opts.force:
exit('WARNING: exact same job already computed, see JOBs table above')
def check_options(opts):
if not opts.workdir: raise Exception('ERROR: output option required.')
# check resume
if not path.exists(opts.workdir) and opts.resume:
print('WARNING: can use output files, found, not resuming...')
opts.resume = False
# sort filters
if opts.apply:
opts.apply.sort()
# check if job already run using md5 digestion of parameters
if already_run(opts) and not opts.force:
exit('WARNING: exact same job already computed, see JOBs table above')
def check_options(opts):
mkdir(opts.workdir)
# transform filtering reads option
opts.filter = filters_to_bin(opts.filter)
# check custom normalization
if opts.normalization == 'custom':
if not opts.biases_path:
raise IOError(
'ERROR: biases file required for "custom" normalization.')
elif not path.exists(opts.biases_path):
raise IOError('ERROR: biases not found at path: %s' %
opts.biases_path)
# check resume
if not path.exists(opts.workdir):
raise IOError('ERROR: workdir not found.')
# for lustre file system....
if 'tmpdb' in opts and opts.tmpdb:
dbdir = opts.tmpdb
# tmp file
dbfile = 'trace_%s' % (''.join(
[ascii_letters[int(random() * 52)] for _ in range(10)]))
opts.tmpdb = path.join(dbdir, dbfile)
try:
copyfile(path.join(opts.workdir, 'trace.db'), opts.tmpdb)
except IOError:
pass
# number of cpus
if opts.cpus == 0:
opts.cpus = cpu_count()
else:
opts.cpus = min(opts.cpus, cpu_count())
# check if job already run using md5 digestion of parameters
try:
if already_run(opts):
if 'tmpdb' in opts and opts.tmpdb:
remove(path.join(dbdir, dbfile))
exit(
'WARNING: exact same job already computed, see JOBs table above'
)
except IOError: # new working directory
pass
def check_options(opts):
# check resume
if not path.exists(opts.workdir):
raise IOError('ERROR: %s does not exists' % opts.workdir)
# number of cpus
if opts.cpus == 0:
opts.cpus = cpu_count()
else:
opts.cpus = min(opts.cpus, cpu_count())
if opts.rich_in_A and opts.fasta:
raise Exception('ERROR: should choose one of FASTA or rich_in_A')
# rich_in_A
if opts.fasta:
if opts.rich_in_A:
raise Exception(
('ERROR: if you input a FASTA file, GC content will'
'will be used as "rich in A" metric to infer '
'compartments.'))
opts.rich_in_A = opts.fasta
# N EVs
if opts.ev_index:
if max(opts.ev_index) > opts.n_evs:
warn(
'WARNING: increasing number of calculated eigenvectors to %d, '
'to match the u=input eigenvectors indices' %
max(opts.ev_index))
opts.n_evs = max(opts.ev_index)
# tmp folder
if 'tmpdb' in opts and opts.tmpdb:
dbdir = opts.tmpdb
# tmp file
dbfile = 'trace_%s' % (''.join(
[ascii_letters[int(random() * 52)] for _ in range(10)]))
opts.tmpdb = path.join(dbdir, dbfile)
copyfile(path.join(opts.workdir, 'trace.db'), opts.tmpdb)
if already_run(opts) and not opts.force:
if 'tmpdb' in opts and opts.tmpdb:
remove(path.join(dbdir, dbfile))
exit('WARNING: exact same job already computed, see JOBs table above')
def check_options(opts):
mkdir(opts.workdir)
# transform filtering reads option
opts.filter = filters_to_bin(opts.filter)
# enlight plotting parameter writing
if opts.only_plot:
opts.plot = True
if opts.interactive:
opts.plot = True
opts.only_plot = True
# check resume
if not path.exists(opts.workdir):
raise IOError('ERROR: workdir not found.')
# for lustre file system....
if 'tmpdb' in opts and opts.tmpdb:
dbdir = opts.tmpdb
# tmp file
dbfile = 'trace_%s' % (''.join(
[ascii_letters[int(random() * 52)] for _ in range(10)]))
opts.tmpdb = path.join(dbdir, dbfile)
try:
copyfile(path.join(opts.workdir, 'trace.db'), opts.tmpdb)
except IOError:
pass
# number of cpus
if opts.cpus == 0:
opts.cpus = cpu_count()
else:
opts.cpus = min(opts.cpus, cpu_count())
# check if job already run using md5 digestion of parameters
if already_run(opts):
if not opts.force:
if 'tmpdb' in opts and opts.tmpdb:
remove(path.join(dbdir, dbfile))
exit(
'WARNING: exact same job already computed, see JOBs table above'
)
else:
warn('WARNING: exact same job already computed, overwritting...')
def check_options(opts):
mkdir(opts.workdir)
# for lustre file system....
if 'tmpdb' in opts and opts.tmpdb:
dbdir = opts.tmpdb
# tmp file
dbfile = 'trace_%s' % (''.join(
[ascii_letters[int(random() * 52)] for _ in range(10)]))
opts.tmpdb = path.join(dbdir, dbfile)
try:
copyfile(path.join(opts.workdir, 'trace.db'), opts.tmpdb)
except IOError:
pass
if opts.workdir1:
# tmp file
dbfile1 = 'trace1_%s' % (''.join(
[ascii_letters[int(random() * 52)] for _ in range(10)]))
opts.tmpdb1 = path.join(dbdir, dbfile1)
try:
copyfile(path.join(opts.workdir1, 'trace.db'), opts.tmpdb1)
except IOError:
pass
if opts.workdir2:
# tmp file
dbfile2 = 'trace2_%s' % (''.join(
[ascii_letters[int(random() * 52)] for _ in range(10)]))
opts.tmpdb2 = path.join(dbdir, dbfile2)
try:
copyfile(path.join(opts.workdir2, 'trace.db'), opts.tmpdb2)
except IOError:
pass
else:
opts.tmpdb1 = path.join(opts.workdir1, 'trace.db')
opts.tmpdb2 = path.join(opts.workdir2, 'trace.db')
# check if job already run using md5 digestion of parameters
if already_run(opts):
if 'tmpdb' in opts and opts.tmpdb:
remove(path.join(dbdir, dbfile))
if opts.workdir1:
remove(path.join(dbdir, dbfile1))
if opts.workdir2:
remove(path.join(dbdir, dbfile2))
exit('WARNING: exact same job already computed, see JOBs table above')
def check_options(opts):
# check resume
if not path.exists(opts.workdir):
raise IOError('ERROR: %s does not exists' % opts.workdir)
if 'tmpdb' in opts and opts.tmpdb:
dbdir = opts.tmpdb
# tmp file
dbfile = 'trace_%s' % (''.join([ascii_letters[int(random() * 52)]
for _ in range(10)]))
opts.tmpdb = path.join(dbdir, dbfile)
copyfile(path.join(opts.workdir, 'trace.db'), opts.tmpdb)
if already_run(opts) and not opts.force:
if 'tmpdb' in opts and opts.tmpdb:
remove(path.join(dbdir, dbfile))
exit('WARNING: exact same job already computed, see JOBs table above')
def check_options(opts):
# check resume
if not path.exists(opts.workdir):
raise IOError('ERROR: %s does not exists' % opts.workdir)
if 'tmpdb' in opts and opts.tmpdb:
dbdir = opts.tmpdb
# tmp file
dbfile = 'trace_%s' % (''.join([ascii_letters[int(random() * 52)]
for _ in range(10)]))
opts.tmpdb = path.join(dbdir, dbfile)
copyfile(path.join(opts.workdir, 'trace.db'), opts.tmpdb)
if already_run(opts) and not opts.force:
if 'tmpdb' in opts and opts.tmpdb:
remove(path.join(dbdir, dbfile))
exit('WARNING: exact same job already computed, see JOBs table above')
def check_options(opts):
# check resume
if not path.exists(opts.workdir) and opts.resume:
print ('WARNING: can use output files, found, not resuming...')
opts.resume = False
if 'tmp' in opts and opts.tmp:
dbdir = opts.tmp
# tmp file
dbfile = 'trace_%s' % (''.join([ascii_letters[int(random() * 52)]
for _ in range(10)]))
opts.tmp = path.join(dbdir, dbfile)
copyfile(path.join(opts.workdir, 'trace.db'), opts.tmp)
if already_run(opts) and not opts.force:
if 'tmp' in opts and opts.tmp:
remove(path.join(dbdir, dbfile))
exit('WARNING: exact same job already computed, see JOBs table above')
def check_options(opts):
# number of cpus
if opts.cpus == 0:
opts.cpus = cpu_count()
else:
opts.cpus = min(opts.cpus, cpu_count())
if opts.rich_in_A and opts.fasta:
raise Exception('ERROR: should choose one of FASTA or rich_in_A')
# rich_in_A
if opts.fasta:
if opts.rich_in_A:
raise Exception(('ERROR: if you input a FASTA file, GC content will'
'will be used as "rich in A" metric to infer '
'compartments.'))
opts.rich_in_A = opts.fasta
# N EVs
if opts.ev_index:
if max(opts.ev_index) > opts.n_evs:
warn('WARNING: increasing number of calculated eigenvectors to %d, '
'to match the u=input eigenvectors indices' % max(opts.ev_index))
opts.n_evs = max(opts.ev_index)
# tmp folder
if 'tmpdb' in opts and opts.tmpdb:
dbdir = opts.tmpdb
# tmp file
dbfile = 'trace_%s' % (''.join([ascii_letters[int(random() * 52)]
for _ in range(10)]))
opts.tmpdb = path.join(dbdir, dbfile)
copyfile(path.join(opts.workdir, 'trace.db'), opts.tmpdb)
try:
_already_run = already_run(opts)
except IOError:
_already_run = False
if _already_run and not opts.force:
if 'tmpdb' in opts and opts.tmpdb:
remove(path.join(dbdir, dbfile))
exit('WARNING: exact same job already computed, see JOBs table above')
def check_options(opts):
mkdir(opts.workdir)
# transform filtering reads option
opts.filter = filters_to_bin(opts.filter)
# check custom normalization
if opts.normalization=='custom':
if not opts.biases_path:
raise IOError('ERROR: biases file required for "custom" normalization.')
elif not path.exists(opts.biases_path):
raise IOError('ERROR: biases not found at path: %s' % opts.biases_path)
# check resume
if not path.exists(opts.workdir):
raise IOError('ERROR: workdir not found.')
# for lustre file system....
if 'tmpdb' in opts and opts.tmpdb:
dbdir = opts.tmpdb
# tmp file
dbfile = 'trace_%s' % (''.join([ascii_letters[int(random() * 52)]
for _ in range(10)]))
opts.tmpdb = path.join(dbdir, dbfile)
try:
copyfile(path.join(opts.workdir, 'trace.db'), opts.tmpdb)
except IOError:
pass
# number of cpus
if opts.cpus == 0:
opts.cpus = cpu_count()
else:
opts.cpus = min(opts.cpus, cpu_count())
# check if job already run using md5 digestion of parameters
try:
if already_run(opts):
if 'tmpdb' in opts and opts.tmpdb:
remove(path.join(dbdir, dbfile))
exit('WARNING: exact same job already computed, see JOBs table above')
except IOError: # new working directory
pass
def check_options(opts):
mkdir(opts.workdir)
# for lustre file system....
if 'tmpdb' in opts and opts.tmpdb:
dbdir = opts.tmpdb
# tmp file
dbfile = 'trace_%s' % (''.join([ascii_letters[int(random() * 52)]
for _ in range(10)]))
opts.tmpdb = path.join(dbdir, dbfile)
try:
copyfile(path.join(opts.workdir, 'trace.db'), opts.tmpdb)
except IOError:
pass
if opts.workdir1:
# tmp file
dbfile1 = 'trace1_%s' % (''.join([ascii_letters[int(random() * 52)]
for _ in range(10)]))
opts.tmpdb1 = path.join(dbdir, dbfile1)
try:
copyfile(path.join(opts.workdir1, 'trace.db'), opts.tmpdb1)
except IOError:
pass
if opts.workdir2:
# tmp file
dbfile2 = 'trace2_%s' % (''.join([ascii_letters[int(random() * 52)]
for _ in range(10)]))
opts.tmpdb2 = path.join(dbdir, dbfile2)
try:
copyfile(path.join(opts.workdir2, 'trace.db'), opts.tmpdb2)
except IOError:
pass
# check if job already run using md5 digestion of parameters
if already_run(opts):
if 'tmpdb' in opts and opts.tmpdb:
remove(path.join(dbdir, dbfile))
if opts.workdir1:
remove(path.join(dbdir, dbfile1))
if opts.workdir2:
remove(path.join(dbdir, dbfile2))
exit('WARNING: exact same job already computed, see JOBs table above')
def check_options(opts):
# check resume
if not path.exists(opts.workdir):
raise IOError('ERROR: wordir not found.')
# for lustre file system....
if 'tmpdb' in opts and opts.tmpdb:
dbdir = opts.tmpdb
# tmp file
dbfile = 'trace_%s' % (''.join(
[ascii_letters[int(random() * 52)] for _ in range(10)]))
opts.tmpdb = path.join(dbdir, dbfile)
try:
copyfile(path.join(opts.workdir, 'trace.db'), opts.tmpdb)
except IOError:
pass
# check if job already run using md5 digestion of parameters
if already_run(opts):
if 'tmpdb' in opts and opts.tmpdb:
remove(path.join(dbdir, dbfile))
exit('WARNING: exact same job already computed, see JOBs table above')
def check_options(opts):
# check resume
if not path.exists(opts.workdir):
raise IOError('ERROR: wordir not found.')
# for lustre file system....
if 'tmpdb' in opts and opts.tmpdb:
dbdir = opts.tmpdb
# tmp file
dbfile = 'trace_%s' % (''.join([ascii_letters[int(random() * 52)]
for _ in range(10)]))
opts.tmpdb = path.join(dbdir, dbfile)
try:
copyfile(path.join(opts.workdir, 'trace.db'), opts.tmpdb)
except IOError:
pass
# check if job already run using md5 digestion of parameters
if already_run(opts):
if 'tmpdb' in opts and opts.tmpdb:
remove(path.join(dbdir, dbfile))
exit('WARNING: exact same job already computed, see JOBs table above')
def check_options(opts):
if not opts.workdir: raise Exception('ERROR: output option required.')
if opts.type != 'map':
raise NotImplementedError('ERROR: not yet there')
if not opts.genome: raise Exception('ERROR: genome parameter required.')
if not opts.workdir: raise Exception('ERROR: workdir parameter required.')
# check skip
if not path.exists(opts.workdir) and opts.skip:
print('WARNING: can use output files, found, not skipping...')
opts.skip = False
if opts.workdir.endswith('/'):
opts.workdir = opts.workdir[:-1]
# write log
newbie = False
if not path.exists(opts.workdir):
newbie = True
mkdir(opts.workdir)
log_format = '[PARSING] %(message)s'
# reset logging
logging.getLogger().handlers = []
try:
print('Writing log to ' + path.join(opts.workdir, 'process.log'))
logging.basicConfig(level=logging.INFO,
format=log_format,
filename=path.join(opts.workdir, 'process.log'),
filemode='a+')
except IOError:
logging.basicConfig(level=logging.DEBUG,
format=log_format,
filename=path.join(opts.workdir, 'process.log2'),
filemode='a+')
# to display log on stdout also
logging.getLogger().addHandler(logging.StreamHandler())
# write version log
vlog_path = path.join(opts.workdir, 'TADbit_and_dependencies_versions.log')
dependencies = get_dependencies_version()
if not path.exists(
vlog_path) or open(vlog_path).readlines() != dependencies:
logging.info('Writing versions of TADbit and dependencies')
vlog = open(vlog_path, 'w')
vlog.write(dependencies)
vlog.close()
# for lustre file system....
if 'tmpdb' in opts and opts.tmpdb:
dbdir = opts.tmpdb
# tmp file
dbfile = 'trace_%s' % (''.join(
[ascii_letters[int(random() * 52)] for _ in range(10)]))
opts.tmpdb = path.join(dbdir, dbfile)
try:
copyfile(path.join(opts.workdir, 'trace.db'), opts.tmpdb)
except IOError:
pass
# check if job already run using md5 digestion of parameters
try:
if already_run(opts):
if 'tmpdb' in opts and opts.tmpdb:
remove(path.join(dbdir, dbfile))
exit(
'WARNING: exact same job already computed, see JOBs table above'
)
except OSError:
pass
def check_options(opts):
if not opts.mapper_binary:
if opts.mapper == 'gem':
opts.mapper_binary = 'gem-mapper'
else:
opts.mapper_binary = opts.mapper
opts.mapper_binary = which(opts.mapper_binary)
if not opts.mapper_binary:
raise Exception(
'\n\nERROR: Mapper binary not found, for GEM install it from:'
'\nhttps://sourceforge.net/projects/gemlibrary/files/gem-library/Binary%20pre-release%202/'
'\n - Download the GEM-binaries-Linux-x86_64-core_i3 if'
'have a recent computer, the '
'GEM-binaries-Linux-x86_64-core_2 otherwise\n - '
'Uncompress with "tar xjvf GEM-binaries-xxx.tbz2"\n - '
'Copy the binary gem-mapper to /usr/local/bin/ for '
'example (somewhere in your PATH).\n\nNOTE: GEM does '
'not provide any binary for MAC-OS.')
opts.gem_version = 0
if opts.mapper == 'gem':
opts.gem_version = None
try:
out, _ = Popen([opts.mapper_binary, '--version'],
stdout=PIPE,
stderr=STDOUT,
universal_newlines=True).communicate()
opts.gem_version = int(out[1])
except ValueError as e:
opts.gem_version = 2
print('Falling to gem v2')
if opts.fast_fragment:
if opts.gem_version < 3:
raise Exception('ERROR: Fast fragment mapping needs GEM v3')
if not opts.fastq2 or not path.exists(opts.fastq2):
raise Exception(
'ERROR: Fast fragment mapping needs both fastq files. '
'Please specify --fastq2')
if opts.read != 0:
raise Exception(
'ERROR: Fast fragment mapping needs to be specified with --read 0'
)
if not opts.genome:
raise Exception('ERROR: Fast fragment mapping needs '
'the genome parameter.')
# check RE name
if opts.renz == ['CHECK']:
print('\nSearching for most probable restriction enzyme in file: %s' %
(opts.fastq))
try:
pat, enz, pv = identify_re(opts.fastq, nreads=100000)
print(' -> Most probable digested site: %s (pv: %f)' % (pat, pv))
print(' -> Enzymes matching: %s' % (', '.join(enz)))
except ValueError:
print(' -> Nothing found...')
exit()
for n, renz in enumerate(opts.renz):
if renz == 'NONE':
opts.renz[n] = None
continue
try:
_ = RESTRICTION_ENZYMES[renz]
except KeyError:
print('\n\nERROR: restriction enzyme %s not found.' % (renz) +
'Use one of:\n\n' + ' '.join(sorted(RESTRICTION_ENZYMES)) +
'\n\n')
raise KeyError()
except AttributeError:
pass
# check skip
if not path.exists(opts.workdir) and opts.skip:
print('WARNING: can use output files, found, not skipping...')
opts.skip = False
# number of cpus
if opts.cpus == 0:
opts.cpus = cpu_count()
else:
opts.cpus = min(opts.cpus, cpu_count())
# check paths
if opts.mapper == 'gem' and not path.exists(opts.index):
raise IOError('ERROR: index file not found at ' + opts.index)
if not path.exists(opts.fastq):
raise IOError('ERROR: FASTQ file not found at ' + opts.fastq)
if not is_fastq(opts.fastq):
raise IOError(
('ERROR: FASTQ file %s wrong format, check') % (opts.fastq))
try:
opts.windows = [[int(i) for i in win.split(':')]
for win in opts.windows]
except TypeError:
pass
mkdir(opts.workdir)
# write log
# if opts.mapping_only:
log_format = '[MAPPING {} READ{}] %(message)s'.format(
opts.fastq, opts.read)
# else:
# log_format = '[DEFAULT] %(message)s'
# reset logging
logging.getLogger().handlers = []
try:
print('Writing log to ' + path.join(opts.workdir, 'process.log'))
logging.basicConfig(level=logging.INFO,
format=log_format,
filename=path.join(opts.workdir, 'process.log'),
filemode='a+')
except IOError:
logging.basicConfig(level=logging.DEBUG,
format=log_format,
filename=path.join(opts.workdir, 'process.log2'),
filemode='a+')
# to display log on stdout also
logging.getLogger().addHandler(logging.StreamHandler())
# write version log
vlog_path = path.join(opts.workdir, 'TADbit_and_dependencies_versions.log')
dependencies = get_dependencies_version()
if not path.exists(
vlog_path) or open(vlog_path).readlines() != dependencies:
logging.info('Writing versions of TADbit and dependencies')
vlog = open(vlog_path, 'w')
vlog.write(dependencies)
vlog.close()
# check mapper extra options
if opts.mapper_param:
if (len(opts.mapper_param) == 1 and
('-' in opts.mapper_param[0] or '--' in opts.mapper_param[0])):
# Single string surrounded by quotes
opts.mapper_param = opts.mapper_param[0].split()
else:
opts.mapper_param = dict([o.split(':') for o in opts.mapper_param])
else:
opts.mapper_param = {}
if opts.mapper == 'gem' and opts.gem_version < 3:
gem_valid_option = set([
"granularity", "q", "quality-format", "gem-quality-threshold",
"mismatch-alphabet", "m", "e", "min-matched-bases",
"max-big-indel-length", "s", "strata-after-best", "fast-mapping",
"unique-mapping", "d", "D", "allow-incomplete-strata",
"max-decoded-matches", "min-decoded-strata", "p",
"paired-end-alignment", "b", "map-both-ends", "min-insert-size",
"max-insert-size", "E", "max-extendable-matches",
"max-extensions-per-match", "unique-pairing"
])
for k in opts.mapper_param:
if not k in gem_valid_option:
raise NotImplementedError(
('ERROR: option "%s" not a valid GEM option'
'or not suported by this tool.') % k)
# create empty DB if don't exists
dbpath = path.join(opts.workdir, 'trace.db')
open(dbpath, 'a').close()
# for lustre file system....
if 'tmpdb' in opts and opts.tmpdb:
dbdir = opts.tmpdb
# tmp file
dbfile = 'trace_%s' % (''.join(
[ascii_letters[int(random() * 52)] for _ in range(10)]))
opts.tmpdb = path.join(dbdir, dbfile)
try:
copyfile(path.join(opts.workdir, 'trace.db'), opts.tmpdb)
except IOError:
pass
# check if job already run using md5 digestion of parameters
if already_run(opts):
if 'tmpdb' in opts and opts.tmpdb:
remove(path.join(dbdir, dbfile))
exit('WARNING: exact same job already computed, see JOBs table above')
def check_options(opts):
if opts.cfg:
get_options_from_cfg(opts.cfg, opts)
opts.gem_binary = which(opts.gem_binary)
if not opts.gem_binary:
raise Exception('\n\nERROR: GEM binary not found, install it from:'
'\nhttps://sourceforge.net/projects/gemlibrary/files/gem-library/Binary%20pre-release%202/'
'\n - Download the GEM-binaries-Linux-x86_64-core_i3 if'
'have a recent computer, the '
'GEM-binaries-Linux-x86_64-core_2 otherwise\n - '
'Uncompress with "tar xjvf GEM-binaries-xxx.tbz2"\n - '
'Copy the binary gem-mapper to /usr/local/bin/ for '
'example (somewhere in your PATH).\n\nNOTE: GEM does '
'not provide any binary for MAC-OS.')
# check RE name
try:
_ = RESTRICTION_ENZYMES[opts.renz]
except KeyError:
print ('\n\nERROR: restriction enzyme not found. Use one of:\n\n'
+ ' '.join(sorted(RESTRICTION_ENZYMES)) + '\n\n')
raise KeyError()
except AttributeError:
pass
# check skip
if not path.exists(opts.workdir) and opts.skip:
print ('WARNING: can use output files, found, not skipping...')
opts.skip = False
# number of cpus
if opts.cpus == 0:
opts.cpus = cpu_count()
else:
opts.cpus = min(opts.cpus, cpu_count())
# check paths
if not path.exists(opts.index):
raise IOError('ERROR: index file not found at ' + opts.index)
if not path.exists(opts.fastq):
raise IOError('ERROR: FASTQ file not found at ' + opts.fastq)
# create tmp directory
if not opts.tmp:
opts.tmp = opts.workdir + '_tmp_r%d' % opts.read
try:
opts.windows = [[int(i) for i in win.split(':')]
for win in opts.windows]
except TypeError:
pass
mkdir(opts.workdir)
# write log
# if opts.mapping_only:
log_format = '[MAPPING {} READ{}] %(message)s'.format(opts.fastq, opts.read)
# else:
# log_format = '[DEFAULT] %(message)s'
# reset logging
logging.getLogger().handlers = []
try:
print 'Writing log to ' + path.join(opts.workdir, 'process.log')
logging.basicConfig(level=logging.INFO,
format=log_format,
filename=path.join(opts.workdir, 'process.log'),
filemode='aw')
except IOError:
logging.basicConfig(level=logging.DEBUG,
format=log_format,
filename=path.join(opts.workdir, 'process.log2'),
filemode='aw')
# to display log on stdout also
logging.getLogger().addHandler(logging.StreamHandler())
# write version log
vlog_path = path.join(opts.workdir, 'TADbit_and_dependencies_versions.log')
dependencies = get_dependencies_version()
if not path.exists(vlog_path) or open(vlog_path).readlines() != dependencies:
logging.info('Writing versions of TADbit and dependencies')
vlog = open(vlog_path, 'w')
vlog.write(dependencies)
vlog.close()
# check GEM mapper extra options
if opts.gem_param:
opts.gem_param = dict([o.split(':') for o in opts.gem_param])
else:
opts.gem_param = {}
gem_valid_option = set(["granularity", "q", "quality-format",
"gem-quality-threshold", "mismatch-alphabet",
"m", "e", "min-matched-bases",
"max-big-indel-length", "s", "strata-after-best",
"fast-mapping", "unique-mapping", "d", "D",
"allow-incomplete-strata", "max-decoded-matches",
"min-decoded-strata", "p", "paired-end-alignment",
"b", "map-both-ends", "min-insert-size",
"max-insert-size", "E", "max-extendable-matches",
"max-extensions-per-match", "unique-pairing"])
for k in opts.gem_param:
if not k in gem_valid_option:
raise NotImplementedError(('ERROR: option "%s" not a valid GEM option'
'or not suported by this tool.') % k)
# check if job already run using md5 digestion of parameters
if already_run(opts):
exit('WARNING: exact same job already computed, see JOBs table above')
