def test_basic_construction(tmpdir, from_file, bundled_piface):
""" PipelineInterface constructor handles Mapping or filepath. """
if from_file:
pipe_iface_config = tmpdir.join("pipe-iface-conf.yaml").strpath
with open(tmpdir.join("pipe-iface-conf.yaml").strpath, 'w') as f:
yaml.safe_dump(bundled_piface, f)
else:
pipe_iface_config = bundled_piface
pi = PipelineInterface(pipe_iface_config)
# Check for the protocol mapping and pipeline interface keys.
assert PL_KEY in pi, "Missing pipeline key ({})".format(PL_KEY)
assert PROTOMAP_KEY in pi, \
"Missing protocol mapping key: ({})".format(PROTOMAP_KEY)
assert pi.pipe_iface_file == (pipe_iface_config if from_file else None)
if from_file:
assert pi.pipelines_path == tmpdir.strpath
else:
assert pi.pipelines_path is None
# Validate protocol mapping and interfaces contents.
assert AttributeDict(bundled_piface[PL_KEY]) == pi[PL_KEY]
assert AttributeDict(bundled_piface[PROTOMAP_KEY]) == pi[PROTOMAP_KEY]
# Certain access modes should agree with one another.
assert pi.pipelines == pi[PL_KEY]
assert list(pi.pipelines.keys()) == pi.pipeline_names
def main():
# Parse command-line arguments
parser = ArgumentParser(prog="starrseq-pipeline",
description="STARR-seq pipeline.")
parser = arg_parser(parser)
parser = pypiper.add_pypiper_args(
parser, groups=["ngs", "looper", "resource", "pypiper"])
args = parser.parse_args()
# Read in yaml configs
sample = STARRSeqSample(pd.Series(yaml.load(open(args.sample_config,
"r"))))
# Check if merged
if len(sample.data_path.split(" ")) > 1:
sample.merged = True
else:
sample.merged = False
sample.prj = AttributeDict(sample.prj)
sample.paths = AttributeDict(sample.paths.__dict__)
# Check read type if not provided
if not hasattr(sample, "ngs_inputs"):
sample.ngs_inputs = [sample.data_source]
if not hasattr(sample, "read_type"):
sample.set_read_type()
# Shorthand for read_type
if sample.read_type == "paired":
sample.paired = True
else:
sample.paired = False
# Set file paths
sample.set_file_paths()
# sample.make_sample_dirs() # should be fixed to check if values of paths are strings and paths indeed
# Start Pypiper object
# Best practice is to name the pipeline with the name of the script;
# or put the name in the pipeline interface.
pipe_manager = pypiper.PipelineManager(name="starrseq",
outfolder=sample.paths.sample_root,
args=args)
pipe_manager.config.tools.scripts_dir = os.path.join(
os.path.dirname(os.path.realpath(__file__)), "tools")
# Start main function
process(sample, pipe_manager, args)
def main():
# Parse command-line arguments
parser = ArgumentParser(
prog="dropseq-pipeline",
description="Drop-seq pipeline."
)
parser = arg_parser(parser)
parser = pypiper.add_pypiper_args(parser, groups=["ngs", "looper", "resource", "pypiper"])
args = parser.parse_args()
if args.sample_config is None:
parser.print_help()
return 1
# Read in yaml configs
sample = AttributeDict(yaml.load(open(args.sample_config, "r")))
pipeline_config = AttributeDict(yaml.load(open(os.path.join(os.path.dirname(__file__), args.config_file), "r")))
# Start main function
process(sample, pipeline_config, args)
def main():
# Parse command-line arguments
parser = ArgumentParser(prog="chipseq-pipeline",
description="ChIP-seq pipeline.")
parser = arg_parser(parser)
parser = pypiper.add_pypiper_args(parser, groups=["all"])
args = parser.parse_args()
if args.sample_config is None:
parser.print_help()
return 1
# Read in yaml configs
series = pd.Series(yaml.load(open(args.sample_config, "r")))
# looper 0.6/0.7 compatibility:
if "protocol" in series.index:
key = "protocol"
elif "library" in series.index:
key = "library"
else:
raise KeyError(
"Sample does not contain either a 'protocol' or 'library' attribute!"
)
# Create Sample object
if series[key] != "ChIPmentation":
sample = ChIPseqSample(series)
else:
sample = ChIPmentation(series)
# Check if merged
if len(sample.data_path.split(" ")) > 1:
sample.merged = True
else:
sample.merged = False
sample.prj = AttributeDict(sample.prj)
sample.paths = AttributeDict(sample.paths.__dict__)
# Check read type if not provided
if not hasattr(sample, "ngs_inputs"):
sample.ngs_inputs = [sample.data_source]
if not hasattr(sample, "read_type"):
sample.set_read_type()
else:
if sample.read_type not in ['single', 'paired']:
sample.set_read_type()
# Shorthand for read_type
if sample.read_type == "paired":
sample.paired = True
else:
sample.paired = False
# Set file paths
sample.set_file_paths()
# sample.make_sample_dirs() # should be fixed to check if values of paths are strings and paths indeed
# Start Pypiper object
# Best practice is to name the pipeline with the name of the script;
# or put the name in the pipeline interface.
pipe_manager = pypiper.PipelineManager(name="chipseq",
outfolder=sample.paths.sample_root,
args=args)
# Start main function
if not args.only_peaks:
pipe_manager = process(sample, pipe_manager, args)
else:
print("Skipped processing sample '{}'.".format(sample.name))
# If sample does not have "ctrl" attribute, finish processing it.
if not hasattr(sample, "compare_sample"):
pipe_manager.stop_pipeline()
print("Finished processing sample '{}'.".format(sample.name))
return
# If compare_sample is empty string, finish processing.
if sample.compare_sample == "":
pipe_manager.stop_pipeline()
print("Finished processing sample '{}'.".format(sample.name))
return
# The pipeline will now wait for the comparison sample file to be completed
pipe_manager._wait_for_file(
sample.filtered.replace(sample.name, sample.compare_sample))
# Start peak calling function
call_peaks(sample, pipe_manager, args)
def main():
# Parse command-line arguments
parser = ArgumentParser(prog="atacseq-pipeline",
description="ATAC-seq pipeline.")
parser = arg_parser(parser)
parser = pypiper.add_pypiper_args(
parser, groups=["ngs", "looper", "resource", "pypiper"])
args = parser.parse_args()
# Read in yaml configs
series = pd.Series(yaml.load(open(args.sample_config, "r")))
# looper 0.6/0.7 compatibility:
if "protocol" in series.index:
key = "protocol"
elif "library" in series.index:
key = "library"
else:
raise KeyError(
"Sample does not contain either a 'protocol' or 'library' attribute!"
)
# Create Sample object
if series[key] != "DNase-seq":
sample = ATACseqSample(series)
else:
sample = DNaseSample(series)
# Check if merged
if len(sample.data_path.split(" ")) > 1:
sample.merged = True
else:
sample.merged = False
sample.prj = AttributeDict(sample.prj)
sample.paths = AttributeDict(sample.paths.__dict__)
# Check read type if not provided
if not hasattr(sample, "ngs_inputs"):
sample.ngs_inputs = [sample.data_source]
if not hasattr(sample, "read_type"):
sample.set_read_type()
else:
if sample.read_type not in ['single', 'paired']:
sample.set_read_type()
# Shorthand for read_type
if sample.read_type == "paired":
sample.paired = True
else:
sample.paired = False
# Set file paths
sample.set_file_paths()
# sample.make_sample_dirs() # should be fixed to check if values of paths are strings and paths indeed
# Start Pypiper object
# Best practice is to name the pipeline with the name of the script;
# or put the name in the pipeline interface.
pipe_manager = pypiper.PipelineManager(name="atacseq",
outfolder=sample.paths.sample_root,
args=args)
pipe_manager.config.tools.scripts_dir = os.path.join(
os.path.dirname(os.path.realpath(__file__)), "tools")
# Start main function
process(sample, pipe_manager, args)