def handle(self, *args, **options):
if options["job_id"] is None:
logger.error("You must specify a job ID.")
return 1
try:
job_type = ProcessorPipeline[options["job_name"]]
except KeyError:
logger.error("You must specify a valid job name.")
return 1
if job_type is ProcessorPipeline.AFFY_TO_PCL:
affy_to_pcl(options["job_id"])
elif job_type is ProcessorPipeline.TRANSCRIPTOME_INDEX:
build_transcriptome_index(options["job_id"])
elif job_type is ProcessorPipeline.NO_OP:
no_op_processor(options["job_id"])
else:
logger.error(
("A valid job name was specified for job %s with id %d but "
"no processor function is known to run it."),
options["job_name"], options["job_id"])
return 1
return 0
def handle(self, *args, **options):
if options["job_id"] is None:
logger.error("You must specify a job ID.",
job_id=options["job_id"])
sys.exit(1)
try:
job_type = ProcessorPipeline[options["job_name"]]
except KeyError:
logger.error("You must specify a valid job name.",
job_name=options["job_name"],
job_id=options["job_id"])
sys.exit(1)
if job_type is ProcessorPipeline.AFFY_TO_PCL:
from data_refinery_workers.processors.array_express import affy_to_pcl
affy_to_pcl(options["job_id"])
elif job_type is ProcessorPipeline.TRANSCRIPTOME_INDEX_SHORT:
from data_refinery_workers.processors.transcriptome_index import build_transcriptome_index
build_transcriptome_index(options["job_id"], length="short")
elif job_type is ProcessorPipeline.TRANSCRIPTOME_INDEX_LONG:
from data_refinery_workers.processors.transcriptome_index import build_transcriptome_index
build_transcriptome_index(options["job_id"], length="long")
elif job_type is ProcessorPipeline.AGILENT_TWOCOLOR_TO_PCL:
from data_refinery_workers.processors.agilent_twocolor import agilent_twocolor_to_pcl
agilent_twocolor_to_pcl(options["job_id"])
elif job_type is ProcessorPipeline.ILLUMINA_TO_PCL:
from data_refinery_workers.processors.illumina import illumina_to_pcl
illumina_to_pcl(options["job_id"])
elif job_type is ProcessorPipeline.SALMON:
from data_refinery_workers.processors.salmon import salmon
salmon(options["job_id"])
elif job_type is ProcessorPipeline.SMASHER:
from data_refinery_workers.processors.smasher import smash
smash(options["job_id"])
elif job_type is ProcessorPipeline.NO_OP:
from data_refinery_workers.processors.no_op import no_op_processor
no_op_processor(options["job_id"])
elif job_type is ProcessorPipeline.JANITOR:
from data_refinery_workers.processors.janitor import run_janitor
run_janitor(options["job_id"])
elif job_type is ProcessorPipeline.QN_REFERENCE:
from data_refinery_workers.processors import qn_reference
qn_reference.create_qn_reference(options["job_id"])
else:
logger.error(
("A valid job name was specified for job %s with id %d but "
"no processor function is known to run it."),
options["job_name"], options["job_id"])
sys.exit(1)
sys.exit(0)
def test_affy_to_pcl_huex_v1(self):
""" Special Case because there is no CDL for Huex V1 """
job = prepare_huex_v1_job()
shutil.rmtree("/home/user/data_store/processor_job_" + str(job.id),
ignore_errors=True)
array_express.affy_to_pcl(job.pk)
updated_job = ProcessorJob.objects.get(pk=job.pk)
self.assertTrue(updated_job.success)
self.assertEqual(len(ComputationalResult.objects.all()), 1)
self.assertEqual(len(ComputedFile.objects.all()), 1)
self.assertEqual(ComputedFile.objects.all()[0].filename,
"GSM1364667_U_110208_7-02-10_S2.PCL")
os.remove(ComputedFile.objects.all()[0].absolute_file_path)
def test_affy_to_pcl_no_brainarray(self):
""" """
job = prepare_non_ba_job()
array_express.affy_to_pcl(job.pk)
updated_job = ProcessorJob.objects.get(pk=job.pk)
self.assertTrue(updated_job.success)
self.assertEqual(len(ComputationalResult.objects.all()), 1)
self.assertEqual(len(ComputedFile.objects.all()), 1)
self.assertEqual(ComputedFile.objects.all()[0].filename,
'GSM45588.PCL')
os.remove(ComputedFile.objects.all()[0].absolute_file_path)
ComputationalResult.objects.all()[0].delete(
) # ComputedFile deleted by cascade
def test_affy_to_pcl(self):
""" """
job = prepare_ba_job()
# Make sure that a previous test didn't leave a directory around.
shutil.rmtree("/home/user/data_store/processor_job_" + str(job.id), ignore_errors=True)
job_context = array_express.affy_to_pcl(job.pk)
self.assertEqual(job_context["platform_accession_code"], "hugene10st")
self.assertEqual(job_context["brainarray_package"], "hugene10sthsensgprobe")
updated_job = ProcessorJob.objects.get(pk=job.pk)
self.assertTrue(updated_job.success)
self.assertEqual(len(ComputationalResult.objects.all()), 1)
self.assertEqual(len(ComputedFile.objects.all()), 1)
self.assertEqual(ComputedFile.objects.all()[0].filename, "GSM1426071_CD_colon_active_1.PCL")
output_filename = ComputedFile.objects.all()[0].absolute_file_path
expected_data = pd.read_csv(
"/home/user/data_store/TEST/PCL/GSM1426071_CD_colon_active_1.PCL", sep="\t"
)["GSM1426071_CD_colon_active_1.CEL"]
actual_data = pd.read_csv(output_filename, sep="\t")["GSM1426071_CD_colon_active_1.CEL"]
assertMostlyAgrees(self, expected_data, actual_data)
os.remove(output_filename)
def test_affy_to_pcl(self):
""" """
job = prepare_ba_job()
# Make sure that a previous test didn't leave a directory around.
shutil.rmtree("/home/user/data_store/processor_job_" + str(job.id),
ignore_errors=True)
array_express.affy_to_pcl(job.pk)
updated_job = ProcessorJob.objects.get(pk=job.pk)
self.assertTrue(updated_job.success)
self.assertEqual(len(ComputationalResult.objects.all()), 1)
self.assertEqual(len(ComputedFile.objects.all()), 1)
self.assertEqual(ComputedFile.objects.all()[0].filename,
"GSM1426071_CD_colon_active_1.PCL")
os.remove(ComputedFile.objects.all()[0].absolute_file_path)
def test_affy_to_pcl_no_brainarray(self):
""" """
job = prepare_non_ba_job()
# Make sure that a previous test didn't leave a directory around.
shutil.rmtree("/home/user/data_store/processor_job_" + str(job.id),
ignore_errors=True)
array_express.affy_to_pcl(job.pk)
updated_job = ProcessorJob.objects.get(pk=job.pk)
self.assertTrue(updated_job.success)
self.assertEqual(len(ComputationalResult.objects.all()), 1)
self.assertEqual(len(ComputedFile.objects.all()), 1)
self.assertEqual(ComputedFile.objects.all()[0].filename,
'GSM45588.PCL')
os.remove(ComputedFile.objects.all()[0].absolute_file_path)
ComputationalResult.objects.all()[0].delete(
) # ComputedFile deleted by cascade
def test_affy_to_pcl_no_brainarray(self):
""" """
job = prepare_non_ba_job()
# Make sure that a previous test didn't leave a directory around.
shutil.rmtree("/home/user/data_store/processor_job_" + str(job.id), ignore_errors=True)
array_express.affy_to_pcl(job.pk)
updated_job = ProcessorJob.objects.get(pk=job.pk)
self.assertTrue(updated_job.success)
self.assertEqual(len(ComputationalResult.objects.all()), 1)
self.assertEqual(len(ComputedFile.objects.all()), 1)
self.assertEqual(ComputedFile.objects.all()[0].filename, "GSM45588.PCL")
output_filename = ComputedFile.objects.all()[0].absolute_file_path
expected_data = pd.read_csv("/home/user/data_store/TEST/PCL/GSM45588.PCL", sep="\t")[
"GSM45588.CEL"
]
actual_data = pd.read_csv(output_filename, sep="\t")["GSM45588.CEL"]
assertMostlyAgrees(self, expected_data, actual_data)
os.remove(output_filename)
