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_salmon_dotsra_bad(self):
try:
os.remove("/home/user/data_store/raw/TEST/SALMON/processed/quant.sf")
except FileNotFoundError:
pass
job, files = prepare_dotsra_job("i-dont-exist.sra")
job_context = salmon.salmon(job.pk)
job = ProcessorJob.objects.get(id=job.pk)
self.assertFalse(job.success)
def test_salmon_dotsra(self):
"""Test the whole pipeline."""
# Ensure any computed files from previous tests are removed.
try:
os.remove("/home/user/data_store/raw/TEST/SALMON/processed/quant.sf")
except FileNotFoundError:
pass
job, files = prepare_dotsra_job()
job_context = salmon.salmon(job.pk)
job = ProcessorJob.objects.get(id=job.pk)
self.assertTrue(job.success)
shutil.rmtree(job_context["work_dir"])
def test_salmon(self):
"""Test the whole pipeline."""
# Ensure any computed files from previous tests are removed.
try:
os.remove("/home/user/data_store/raw/TEST/SALMON/processed/quant.sf")
except FileNotFoundError:
pass
job, files = prepare_job()
job_context = salmon.salmon(job.pk)
job = ProcessorJob.objects.get(id=job.pk)
self.assertTrue(job.success)
sample = files[0].samples.first()
self.assertFalse(sample.is_processed)
organism_index = job_context["quant_result"].organism_index
self.assertEqual(organism_index.index_type, "TRANSCRIPTOME_SHORT")
def test_no_salmon_on_geo(self):
"""Test that salmon won't be run on data coming from GEO."""
# Ensure any computed files from previous tests are removed.
try:
os.remove("/home/user/data_store/raw/TEST/SALMON/processed/quant.sf")
except FileNotFoundError:
pass
job, files = prepare_job()
# We're expecting this processor job to fail, and when it does
# it should clean up the original files that were for the
# job. However we want to use these files in other tests, so
# copy them so we can delete them without deleting the
# originals.
for original_file in OriginalFile.objects.all():
new_path = original_file.absolute_file_path + "_copy"
shutil.copyfile(original_file.absolute_file_path, new_path)
original_file.absolute_file_path = new_path
original_file.save()
sample_object = Sample.objects.first()
sample_object.source_database = 'GEO'
sample_object.save()
job_context = salmon.salmon(job.pk)
job = ProcessorJob.objects.get(id=job.pk)
self.assertFalse(job.success)
self.assertEqual(job.failure_reason,
("The sample for this job either was not RNA-Seq or was not from the "
"SRA database."))
self.assertTrue(job.no_retry)
# Make sure the data got cleaned up, since the Janitor isn't
# going to do it.
for original_file in OriginalFile.objects.all():
self.assertFalse(os.path.exists(original_file.absolute_file_path))