def _run_db2falcon_jobs(cwd, config, dry_run, db2falcon_json_fn=None):
"""
cwd --- current workding directory
dry_run --- if True, do not actually run the scripts
db2falcon_json_fn --- if not None, write dict{0: dict('script_fn':script_fn, 'script_dir':script_dir)}
"""
# Generate preads4falcon.fasta from preads.db
script_fn = os.path.join(cwd, "run_db2falcon.sh")
job_done = script_fn + "_done"
args = {
"config": config,
"job_done": job_done,
"script_fn": script_fn,
"preads4falcon_fn": "preads4falcon.fasta",
"preads_db": "preads.db",
}
support.run_db2falcon(**args)
json_fn = "db2falcon.json" if db2falcon_json_fn is None else db2falcon_json_fn
# add script_dir to args for scattered tasks to work in the correct dir
json_data = {0: {"script_fn": os.path.basename(script_fn), "script_dir": cwd}}
with open(json_fn, "w") as writer:
writer.write(json.dumps(json_data) + "\n")
mkdir(cwd)
with cd(cwd):
if dry_run is False:
run_cmd("bash %s" % os.path.basename(script_fn), sys.stdout, sys.stderr, shell=False)
assert_nonzero("preads4falcon.fasta")
def run_merge_consensus_jobs(input_files, output_files, db_prefix='raw_reads'):
print('run_merge_consensus_jobs: %s %s %s' %
(db_prefix, repr(input_files), repr(output_files)))
i_json_config_fn, run_daligner_job_fn, i_fofn_fn = input_files
o_fofn_fn, = output_files
db_dir = os.path.dirname(run_daligner_job_fn)
cmds = ['pwd', 'ls -al']
fns = ['.{pre}.bps', '.{pre}.idx', '{pre}.db']
cmds += ['rm -f %s' % fn for fn in fns]
cmds += ['ln -sf {dir}/%s .' % fn for fn in fns]
cmd = ';'.join(cmds).format(dir=os.path.relpath(db_dir), pre=db_prefix)
run_cmd(cmd, sys.stdout, sys.stderr, shell=True)
cwd = os.getcwd() # same as dir of o_fofn_fn
config = _get_config_from_json_fileobj(open(i_json_config_fn))
# i_fofn_fn has the .las files, so create_merge_tasks does not need to look for theme.
tasks = create_merge_tasks(i_fofn_fn,
run_daligner_job_fn,
cwd,
db_prefix=db_prefix,
config=config)
las_fns = _run_merge_jobs(
dict((p_id, (argstuple[0], argstuple[2]))
for (p_id, argstuple) in tasks.items()))
if db_prefix == 'raw_reads':
fasta_fns = _run_consensus_jobs(
dict((p_id, (argstuple[1], argstuple[3]))
for (p_id, argstuple) in tasks.items()))
# Record '*.fasta' in FOFN.
write_fns(o_fofn_fn, sorted(os.path.abspath(f) for f in fasta_fns))
assert_nonzero(o_fofn_fn)
return
# Record '*.las' from merge_jobs in FOFN.
write_fns(o_fofn_fn, sorted(os.path.abspath(f) for f in las_fns))
# Generate preads4falcon.fasta from preads.db
script_fn = os.path.join(cwd, "run_db2falcon.sh")
job_done = script_fn + '_done'
args = {
'config': config,
'job_done': job_done,
'script_fn': script_fn,
}
support.run_db2falcon(**args)
run_cmd('bash %s' % script_fn, sys.stdout, sys.stderr, shell=False)
assert_nonzero('preads4falcon.fasta')
assert_nonzero(o_fofn_fn)
return 0
def run_merge_consensus_jobs(input_files, output_files, db_prefix='raw_reads'):
print('run_merge_consensus_jobs: %s %s %s' %(db_prefix, repr(input_files), repr(output_files)))
i_json_config_fn, run_daligner_job_fn, i_fofn_fn = input_files
o_fofn_fn, = output_files
db_dir = os.path.dirname(run_daligner_job_fn)
cmds = ['pwd', 'ls -al']
fns = ['.{pre}.bps', '.{pre}.idx', '{pre}.db']
cmds += ['rm -f %s' %fn for fn in fns]
cmds += ['ln -sf {dir}/%s .' %fn for fn in fns]
cmd = ';'.join(cmds).format(
dir=os.path.relpath(db_dir), pre=db_prefix)
run_cmd(cmd, sys.stdout, sys.stderr, shell=True)
cwd = os.getcwd() # same as dir of o_fofn_fn
config = _get_config_from_json_fileobj(open(i_json_config_fn))
# i_fofn_fn has the .las files, so create_merge_tasks does not need to look for theme.
tasks = create_merge_tasks(i_fofn_fn, run_daligner_job_fn, cwd, db_prefix=db_prefix, config=config)
las_fns = _run_merge_jobs(
dict((p_id, (argstuple[0], argstuple[2])) for (p_id, argstuple) in tasks.items()))
if db_prefix == 'raw_reads':
fasta_fns = _run_consensus_jobs(
dict((p_id, (argstuple[1], argstuple[3])) for (p_id, argstuple) in tasks.items()))
# Record '*.fasta' in FOFN.
write_fns(o_fofn_fn, sorted(os.path.abspath(f) for f in fasta_fns))
assert_nonzero(o_fofn_fn)
return
# Record '*.las' from merge_jobs in FOFN.
write_fns(o_fofn_fn, sorted(os.path.abspath(f) for f in las_fns))
# Generate preads4falcon.fasta from preads.db
script_fn = os.path.join(cwd, "run_db2falcon.sh")
job_done = script_fn + '_done'
args = {
'config': config,
'job_done': job_done,
'script_fn': script_fn,
}
support.run_db2falcon(**args)
run_cmd('bash %s' %script_fn, sys.stdout, sys.stderr, shell=False)
assert_nonzero('preads4falcon.fasta')
assert_nonzero(o_fofn_fn)
def _run_db2falcon_jobs(cwd, config, dry_run, db2falcon_json_fn=None):
"""
cwd --- current workding directory
dry_run --- if True, do not actually run the scripts
db2falcon_json_fn --- if not None, write dict{0: dict('script_fn':script_fn, 'script_dir':script_dir)}
"""
# Generate preads4falcon.fasta from preads.db
script_fn = os.path.join(cwd, "run_db2falcon.sh")
job_done = script_fn + '_done'
args = {
'config': config,
'job_done': job_done,
'script_fn': script_fn,
'preads4falcon_fn': 'preads4falcon.fasta',
'preads_db': 'preads.db',
}
support.run_db2falcon(**args)
json_fn = "db2falcon.json" if db2falcon_json_fn is None else db2falcon_json_fn
# add script_dir to args for scattered tasks to work in the correct dir
json_data = {
0: {
'script_fn': os.path.basename(script_fn),
'script_dir': cwd
}
}
with open(json_fn, 'w') as writer:
writer.write(json.dumps(json_data) + "\n")
mkdir(cwd)
with cd(cwd):
if dry_run is False:
run_cmd('bash %s' % os.path.basename(script_fn),
sys.stdout,
sys.stderr,
shell=False)
assert_nonzero('preads4falcon.fasta')
