def run_workflow(config, jobs, out_dir, no_conda, dryrun, snakemake_args, workflow):
if not os.path.exists(config):
logging.critical("Config not found: %s" % config)
sys.exit(1)
validate_config(config, workflow)
out_dir = os.path.realpath(out_dir)
cmd = ("snakemake --snakefile {snakefile} --directory {out_dir} "
"--printshellcmds --jobs {jobs} --rerun-incomplete "
"--configfile '{config}' --nolock {conda} {dryrun} "
"--config workflow={workflow} {add_args} "
"{args}").format(snakefile=get_snakefile(),
out_dir=out_dir,
jobs=jobs,
config=config,
conda="" if no_conda else "--use-conda",
dryrun="--dryrun" if dryrun else "",
add_args="" if snakemake_args and snakemake_args[0].startswith("-") else "--",
args=" ".join(snakemake_args),
workflow=workflow)
logging.info("Executing: %s" % cmd)
try:
subprocess.check_call(cmd, shell=True)
except subprocess.CalledProcessError as e:
# removes the traceback
logging.critical(e)
def run_workflow(workflow, working_dir, config_file, jobs, profile, dryrun,
snakemake_args):
"""Runs the ATLAS pipline
By default all steps are executed but a sub-workflow can be specified.
Needs a config-file and expects to find a sample table in the working-directory. Both can be generated with 'atlas init'
Most snakemake arguments can be appended to the command for more info see 'snakemake --help'
For more details, see: https://metagenome-atlas.readthedocs.io
"""
if config_file is None:
config_file = os.path.join(working_dir, "config.yaml")
if not os.path.exists(config_file):
logging.critical(f"config-file not found: {config_file}\n"
"generate one with 'atlas init'")
sys.exit(1)
sample_file = os.path.join(working_dir, "samples.tsv")
if not os.path.exists(sample_file):
logging.critical(f"sample.tsv not found in the working directory. "
"Generate one with 'atlas init'")
sys.exit(1)
validate_config(config_file, workflow)
conf = load_configfile(config_file)
db_dir = conf["database_dir"]
cmd = ("snakemake --snakefile {snakefile} --directory {working_dir} "
"{jobs} --rerun-incomplete "
"--configfile '{config_file}' --nolock "
" {profile} --use-conda {conda_prefix} {dryrun} "
" --scheduler greedy "
" {target_rule} "
" {args} ").format(
snakefile=get_snakefile(),
working_dir=working_dir,
jobs="--jobs {}".format(jobs) if jobs is not None else "",
config_file=config_file,
profile="" if
(profile is None) else "--profile {}".format(profile),
dryrun="--dryrun" if dryrun else "",
args=" ".join(snakemake_args),
target_rule=workflow if workflow != "None" else "",
conda_prefix="--conda-prefix " +
os.path.join(db_dir, "conda_envs"),
)
logging.info("Executing: %s" % cmd)
try:
subprocess.check_call(cmd, shell=True)
except subprocess.CalledProcessError as e:
# removes the traceback
logging.critical(e)
exit(1)