def check_reference(meta, args):
"""Check and install the needed reference genome"""
print("Checking reference")
plan_data = json.load(open(os.path.join(args.path, "plan.json")))
version, design, meta = ampliseq.handle_versioned_plans(plan_data, meta)
print("Got versioned stuff")
# If we have a genome reference, check to see if it's installed
reference = design.get('genome_reference', None)
print(reference)
if not reference:
return False
try:
url = reference.get('uri')
ref_hash = reference.get('files_md5sum', {}).get('fasta')
short_name = reference.get('short_name')
name = reference.get('name')
notes = reference.get('notes', "AmpliSeq Import")
print("Got various fields")
except KeyError as err:
# If the key does not exist, that's fine, but it can't exist and be corrupt
print("Corrupt genome_reference entry: {0}".format(err))
sys.exit(1)
# The identity_hash matching the files_md5sum.fasta hash determines whether
# or not the genome is installed
print("Checking reference " + ref_hash)
if not models.ReferenceGenome.objects.filter(identity_hash=ref_hash).exists():
reference = models.ReferenceGenome(
enabled = False,
identity_hash = ref_hash,
name = name,
notes = notes,
short_name = short_name,
source = url,
status = "downloading",
index_version = "tmap-f3"
)
reference.save()
print("created new reference")
pub = models.Publisher.objects.get(name='BED')
upload = models.ContentUpload.objects.get(pk=args.upload_id)
# This is a celery subtask that will run the publisher scripts on this upload again
finish_me = run_pub_scripts.si(pub, upload)
print("About t set check point")
set_checkpoint(meta, args)
print("check point set")
# With a status starting with "Waiting" the framework will stop
# after pre_processing, before validate.
upload.status = "Waiting on reference"
upload.save()
# the subtask finish_me will be called at the end of the reference install
# process to restart validation of the upload
start_reference_download(url, reference, callback=finish_me)
print("Started reference download")
return True
print("exiting in shame")
return False
def pre_process():
parse = argparse.ArgumentParser()
parse.add_argument('upload_id', type=int)
parse.add_argument('path')
parse.add_argument('upload_file')
parse.add_argument('meta_file', type=argparse.FileType('r+'))
try:
args = parse.parse_args()
except IOError as err:
print("Input file error: %s" % err)
parse.print_help()
sys.exit(1)
meta = json.load(args.meta_file)
meta.update({
"is_ampliseq": None,
"primary_bed": None,
"hotspot_bed": None
})
is_zip = zipfile.is_zipfile(args.upload_file)
if is_zip:
files = unzip_archive(args.path, args.upload_file)
else:
files = [args.upload_file]
if len(files) == 1 and files[0].endswith('.bed'):
meta['is_ampliseq'] = False
meta['primary_bed'] = files[0]
elif "plan.json" in files:
print("Found ampliseq")
meta['is_ampliseq'] = True
plan_data = json.load(open(os.path.join(args.path, "plan.json")))
version, design = ampliseq.handle_versioned_plans(plan_data)
meta['design'] = design
plan = design['plan']
try:
meta['primary_bed'] = plan['designed_bed']
meta['secondary_bed'] = plan['hotspot_bed']
if 'reference' not in meta:
meta['reference'] = plan['genome'].lower()
except KeyError as err:
api.patch("contentupload", args.upload_id, status="Error: malformed AmpliSeq archive")
raise
print(meta)
else:
raise ValueError("Upload must be either valid Ampliseq export or contain a single BED file.")
args.meta_file.truncate(0)
args.meta_file.seek(0)
json.dump(meta, args.meta_file)
api.patch("contentupload", args.upload_id, meta=meta)
def get_ampliseq_fixed_designs(user, password):
h = httplib2.Http(disable_ssl_certificate_validation=settings.DEBUG)
h.add_credentials(user, password)
url = urlparse.urljoin(settings.AMPLISEQ_URL, "ws/tmpldesign/list/active")
response, content = h.request(url)
if response['status'] == '200':
designs = json.loads(content)
fixed = []
for template in designs.get('TemplateDesigns', []):
version, data, meta = ampliseq.handle_versioned_plans(template)
fixed.append(data)
return response, fixed
else:
return response, None
def check_reference(meta, args):
print("Checking reference")
plan_data = json.load(open(os.path.join(args.path, "plan.json")))
version, design, meta = ampliseq.handle_versioned_plans(plan_data, meta)
print("Got versioned stuff")
reference = design.get('genome_reference', None)
print(reference)
if not reference:
return False
try:
url = reference.get('uri')
ref_hash = reference.get('files_md5sum', {}).get('fasta')
short_name = reference.get('short_name')
name = reference.get('name')
notes = reference.get('notes', "AmpliSeq Import")
print("Got various fields")
except KeyError as err:
print("Corrupt genome_reference entry: {0}".format(err))
sys.exit(1)
print("Checking reference " + ref_hash)
if not models.ReferenceGenome.objects.filter(identity_hash=ref_hash).exists():
reference = models.ReferenceGenome(
enabled = False,
identity_hash = ref_hash,
name = name,
notes = notes,
short_name = short_name,
source = url,
status = "downloading",
index_version = "tmap-f3"
)
reference.save()
print("created new reference")
pub = models.Publisher.objects.get(name='BED')
upload = models.ContentUpload.objects.get(pk=args.upload_id)
finish_me = run_pub_scripts.si(pub, upload)
print("About t set check point")
set_checkpoint(meta, args)
print("check point set")
upload.status = "Waiting on reference"
upload.save()
start_reference_download(url, reference, callback=finish_me)
print("Started reference download")
return True
print("exiting in shame")
return False
def get_ampliseq_designs(user, password):
h = httplib2.Http(disable_ssl_certificate_validation=settings.DEBUG)
h.add_credentials(user, password)
url = urlparse.urljoin(settings.AMPLISEQ_URL, "ws/design/list")
response, content = h.request(url)
if response['status'] == '200':
design_data = json.loads(content)
designs = design_data.get('AssayDesigns', [])
for design in designs:
solutions = []
for solution in design.get('DesignSolutions', []):
version, data, meta = ampliseq.handle_versioned_plans(solution)
solutions.append(data)
design['DesignSolutions'] = solutions
return response, designs
else:
return response, {}
def check_reference(meta, args):
print("Checking reference")
plan_data = json.load(open(os.path.join(args.path, "plan.json")))
version, design, meta = ampliseq.handle_versioned_plans(plan_data, meta)
print("Got versioned stuff")
reference = design.get('genome_reference', None)
print(reference)
if not reference:
return False
try:
url = reference.get('uri')
ref_hash = reference.get('files_md5sum', {}).get('fasta')
short_name = reference.get('short_name')
name = reference.get('name')
notes = reference.get('notes', "AmpliSeq Import")
print("Got various fields")
except KeyError as err:
print("Corrupt genome_reference entry: {0}".format(err))
sys.exit(1)
print("Checking reference " + ref_hash)
if not models.ReferenceGenome.objects.filter(
identity_hash=ref_hash).exists():
reference = models.ReferenceGenome(enabled=False,
identity_hash=ref_hash,
name=name,
notes=notes,
short_name=short_name,
source=url,
status="downloading",
index_version="tmap-f3")
reference.save()
print("created new reference")
pub = models.Publisher.objects.get(name='BED')
upload = models.ContentUpload.objects.get(pk=args.upload_id)
finish_me = run_pub_scripts.si(pub, upload)
print("About t set check point")
set_checkpoint(meta, args)
print("check point set")
upload.status = "Waiting on reference"
upload.save()
start_reference_download(url, reference, callback=finish_me)
print("Started reference download")
return True
print("exiting in shame")
return False
def pre_process():
parse = argparse.ArgumentParser()
parse.add_argument('upload_id', type=int)
parse.add_argument('path')
parse.add_argument('upload_file')
parse.add_argument('meta_file', type=argparse.FileType('r+'))
try:
args = parse.parse_args()
except IOError as err:
print("Input file error: %s" % err)
parse.print_help()
sys.exit(1)
meta = json.load(args.meta_file, parse_float=Decimal)
meta.update({
"is_ampliseq": None,
"primary_bed": None,
"hotspot_bed": None
})
is_zip = zipfile.is_zipfile(args.upload_file)
if is_zip:
files = unzip_archive(args.path, args.upload_file)
else:
files = [args.upload_file]
if len(files) == 1 and files[0].endswith('.bed'):
meta['is_ampliseq'] = False
meta['primary_bed'] = files[0]
elif len(files) == 1 and files[0].endswith('.vcf') and meta['hotspot']:
# convert vcf to bed
target_filename = os.path.join(args.path,os.path.basename(files[0])) + '.bed'
convert_command = '/usr/local/bin/tvcutils prepare_hotspots'
convert_command += ' --input-vcf %s' % files[0]
convert_command += ' --output-bed %s' % target_filename
convert_command += ' --reference /results/referenceLibrary/tmap-f3/%s/%s.fasta' % (meta["reference"],meta["reference"]) #TODO: is this just name or full path??
convert_command += ' --filter-bypass on'
process = subprocess.Popen(convert_command, stderr=subprocess.STDOUT, stdout=subprocess.PIPE, shell=True)
for line in process.communicate()[0].splitlines():
api.post('log', upload='/rundb/api/v1/contentupload/%s/' % str(args.upload_id), text=line.strip())
meta['is_ampliseq'] = False
meta['primary_bed'] = target_filename
'''
elif len(files) == 1 and files[0].endswith('.vcf') and meta['hotspot']:
# convert vcf to bed
target_filename = os.path.join(args.path,os.path.basename(files[0])) + '.bed'
convert_command = '/usr/local/bin/tvcutils prepare_hotspots'
convert_command += ' --input-vcf %s' % files[0]
convert_command += ' --output-bed %s' % target_filename
convert_command += ' --reference /results/referenceLibrary/tmap-f3/%s/%s.fasta' % (meta["reference"],meta["reference"]) #TODO: is this just name or full path??
convert_command += ' --filter-bypass on'
process = subprocess.Popen(convert_command, stderr=subprocess.STDOUT, stdout=subprocess.PIPE, shell=True)
for line in process.communicate()[0]:
api.post('log', upload='/rundb/api/v1/contentupload/%s/' % str(args.upload_id), text=line.strip())
meta['is_ampliseq'] = False
meta['primary_bed'] = target_filename
'''
elif "plan.json" in files:
print("Found ampliseq")
meta['is_ampliseq'] = True
plan_data = json.load(open(os.path.join(args.path, "plan.json")), parse_float=Decimal)
version, design = ampliseq.handle_versioned_plans(plan_data)
meta['design'] = design
plan = design['plan']
try:
meta['primary_bed'] = plan['designed_bed']
meta['secondary_bed'] = plan['hotspot_bed']
if not meta.get("reference", None):
meta['reference'] = design['genome'].lower()
except KeyError as err:
api.patch("contentupload", args.upload_id, status="Error: malformed AmpliSeq archive")
raise
print(meta)
else:
raise ValueError("Upload must be either valid Ampliseq export or contain a single BED file.")
args.meta_file.truncate(0)
args.meta_file.seek(0)
json.dump(meta, args.meta_file, cls=JSONEncoder)
api.patch("contentupload", args.upload_id, meta=meta)
def main():
parse = argparse.ArgumentParser()
parse.add_argument('upload_id', type=int)
parse.add_argument('path')
parse.add_argument('upload_file')
parse.add_argument('meta_file')
try:
args = parse.parse_args()
except IOError as err:
print("ERROR: Input file error: %s" % err)
parse.print_help()
sys.exit(1)
with open(args.meta_file) as f:
meta = json.load(f, parse_float=Decimal)
files = meta.get('pre_process_files')
target_regions_bed = None
hotspots_bed = None
meta['is_ampliseq'] = False
if len(files) == 1 and files[0].endswith('.bed') and meta.get('hotspot', False) == False:
target_regions_bed = os.path.basename(files[0])
meta['is_ampliseq'] = False
print "Content: Target regions file in BED format"
print
elif len(files) == 1 and files[0].endswith('.bed') and meta.get('hotspot', False) == True:
hotspots_bed = os.path.basename(files[0])
meta['is_ampliseq'] = False
print "Content: Hotspots file in BED format"
print
elif len(files) == 1 and files[0].endswith('.vcf') and meta.get('hotspot', False) == True:
print "Content: Hotspots file in VCF format"
print
print "Converting hotspot VCF file to BED: %s" % files[0]
print
hotspots_bed = os.path.basename(files[0]) + '.bed'
convert_command = '/usr/local/bin/tvcutils prepare_hotspots'
convert_command += ' --input-vcf %s' % os.path.join(args.path, os.path.basename(files[0]))
convert_command += ' --output-bed %s' % os.path.join(args.path, hotspots_bed)
convert_command += ' --reference /results/referenceLibrary/tmap-f3/%s/%s.fasta' % (
meta["reference"], meta["reference"])
convert_command += ' --filter-bypass on'
p = subprocess.Popen(convert_command, stderr=subprocess.STDOUT, stdout=subprocess.PIPE, shell=True)
print p.communicate()[0]
if p.returncode != 0:
sys.exit(p.returncode)
#process = subprocess.Popen(convert_command, stderr=subprocess.STDOUT, stdout=subprocess.PIPE, shell=True)
#for line in process.communicate()[0].splitlines():
# api.post('log', upload='/rundb/api/v1/contentupload/%s/' %
# str(args.upload_id), text=line.strip())
meta['is_ampliseq'] = False
elif "plan.json" in files:
print "Content: AmpliSeq ZIP"
print
meta['is_ampliseq'] = True
plan_data = json.load(open(os.path.join(args.path, "plan.json")))
version, design, meta = ampliseq.handle_versioned_plans(plan_data, meta, args.path)
meta['design'] = design
try:
target_regions_bed = design['plan']['designed_bed']
hotspots_bed = design['plan']['hotspot_bed']
if not meta.get("reference", None):
meta['reference'] = design['genome']
if 'design_name' in plan_data:
meta['description'] = design['design_name']
api.update_meta(meta, args)
except KeyError as err:
api.patch("contentupload", args.upload_id, status="Error: malformed AmpliSeq archive")
print "ERROR: Malformed AmpliSeq archive: missing json key "+str(err)
sys.exit(1)
#api.post('log', upload='/rundb/api/v1/contentupload/%s/' % str(args.upload_id), text="Malformed AmpliSeq archive: missing json key "+str(err))
#raise
if target_regions_bed and target_regions_bed not in files:
api.patch("contentupload", args.upload_id, status="Error: malformed AmpliSeq archive")
print "ERROR: Target region file %s not present in AmpliSeq archive" % target_regions_bed
sys.exit(1)
#api.post('log', upload='/rundb/api/v1/contentupload/%s/' % str(args.upload_id),
# text="Malformed AmpliSeq archive: Target region file %s not present in AmpliSeq archive" % target_regions_bed)
#raise ValueError("Target region file %s not present in AmpliSeq archive" % target_regions_bed)
if hotspots_bed and hotspots_bed not in files:
api.patch("contentupload", args.upload_id, status="Error: malformed AmpliSeq archive")
print "ERROR: Hotspots file %s not present in AmpliSeq archive" % target_regions_bed
sys.exit(1)
#api.post('log', upload='/rundb/api/v1/contentupload/%s/' % str(args.upload_id),
# text="Malformed AmpliSeq archive: Hotspots file %s not present in AmpliSeq archive" % target_regions_bed)
#raise ValueError("Hotspots file %s not present in AmpliSeq archive" % target_regions_bed)
else:
api.patch("contentupload", args.upload_id, status="Error: Unrecognized upload type.")
print
print "ERROR: Unrecognized upload type. Upload must be either a valid Ampliseq ZIP or contain a single BED or VCF file."
sys.exit(1)
''' === Validate and Register === '''
primary_path = None
secondary_path = None
if is_BED_encrypted(meta):
if target_regions_bed:
meta['design']['plan']['designed_bed'] = ''
if hotspots_bed:
meta['design']['plan']['hotspot_bed'] = ''
primary_path = ""
secondary_path = ""
else:
if target_regions_bed:
primary_path = validate(args.upload_id, args.path, meta, target_regions_bed, 'target regions BED')
if hotspots_bed:
secondary_path = validate(args.upload_id, args.path, meta, hotspots_bed, 'hotspots BED')
meta["hotspot"] = False
if target_regions_bed and not primary_path:
register_bed_file(args.upload_id, args.path, meta, target_regions_bed)
if hotspots_bed:
meta["hotspot"] = True
if not secondary_path:
register_bed_file(args.upload_id, args.path, meta, hotspots_bed)
if meta['is_ampliseq']:
try:
if not (is_BED_encrypted(meta)):
if target_regions_bed and not primary_path:
primary_path = os.path.join(
args.path, meta["reference"]+"/unmerged/detail/"+target_regions_bed)
if hotspots_bed and not secondary_path:
secondary_path = os.path.join(
args.path, meta["reference"]+"/unmerged/detail/"+hotspots_bed)
else:
run_type = meta['design']['plan'].get('runType', None)
if run_type and (run_type == "AMPS_RNA"):
meta['reference'] = None
plan_prototype, alignmentargs_override = plan_json(
meta, args.upload_id, primary_path, secondary_path)
success, response, content = api.post("plannedexperiment", **plan_prototype)
if not success:
api.patch("contentupload", args.upload_id, status="Error: unable to create TS Plan")
err_content = json.loads(content)
error_message_array = []
if 'error' in err_content:
error_json = json.loads(str(err_content['error'][3:-2]))
for k in error_json:
for j in range(len(error_json[k])):
err_message = str(error_json[k][j])
err_message = err_message.replace('>', '>')
error_message_array.append(err_message)
error_messages = ','.join(error_message_array)
raise Exception(error_messages)
if alignmentargs_override:
content_dict = json.loads(content)
api.patch("plannedexperiment", content_dict[
"id"], alignmentargs=alignmentargs_override, thumbnailalignmentargs=alignmentargs_override)
except Exception as err:
print("ERROR: Could not create plan from this zip: %s." % err)
raise
api.update_meta(meta, args)
def main():
parse = argparse.ArgumentParser()
parse.add_argument('upload_id', type=int)
parse.add_argument('path')
parse.add_argument('upload_file')
parse.add_argument('meta_file')
try:
args = parse.parse_args()
except IOError as err:
print("Input file error: %s" % err)
parse.print_help()
sys.exit(1)
meta_file_handle = open(args.meta_file,'r')
meta = json.load(meta_file_handle, parse_float=Decimal)
meta_file_handle.close()
is_zip = zipfile.is_zipfile(args.upload_file)
if is_zip:
files = file_utils.unzip_archive(args.path, args.upload_file)
elif args.upload_file.endswith('.gz'):
files = [os.path.basename(args.upload_file[:-3])]
cmd = 'gzip -dc %s > %s ' % (args.upload_file, os.path.join(args.path,files[0]))
subprocess.call(cmd, shell=True)
else:
files = [args.upload_file]
''' Establish the upload type '''
target_regions_bed = None
hotspots_bed = None
meta['is_ampliseq'] = False
if len(files) == 1 and files[0].endswith('.bed') and meta.get('hotspot',False) == False:
target_regions_bed = os.path.basename(files[0])
meta['is_ampliseq'] = False
elif len(files) == 1 and files[0].endswith('.bed') and meta.get('hotspot',False) == True:
hotspots_bed = os.path.basename(files[0])
meta['is_ampliseq'] = False
elif len(files) == 1 and files[0].endswith('.vcf') and meta.get('hotspot',False) == True:
hotspots_bed = os.path.basename(files[0]) + '.bed'
convert_command = '/usr/local/bin/tvcutils prepare_hotspots'
convert_command += ' --input-vcf %s' % os.path.join(args.path,os.path.basename(files[0]))
convert_command += ' --output-bed %s' % os.path.join(args.path,hotspots_bed)
convert_command += ' --reference /results/referenceLibrary/tmap-f3/%s/%s.fasta' % (meta["reference"],meta["reference"]) #TODO: is this just name or full path??
convert_command += ' --filter-bypass on'
process = subprocess.Popen(convert_command, stderr=subprocess.STDOUT, stdout=subprocess.PIPE, shell=True)
for line in process.communicate()[0].splitlines():
api.post('log', upload='/rundb/api/v1/contentupload/%s/' % str(args.upload_id), text=line.strip())
meta['is_ampliseq'] = False
elif "plan.json" in files:
meta['is_ampliseq'] = True
plan_data = json.load(open(os.path.join(args.path, "plan.json")))
version, design = ampliseq.handle_versioned_plans(plan_data, meta)
meta['design'] = design
try:
target_regions_bed = design['plan']['designed_bed']
hotspots_bed = design['plan']['hotspot_bed']
if not meta.get("reference", None):
meta['reference'] = design['genome'].lower()
if 'design_name' in plan_data:
meta['description'] = design['design_name']
meta_file_handle = open(args.meta_file,'w')
json.dump(meta, meta_file_handle, cls=JSONEncoder)
meta_file_handle.close()
api.patch("contentupload", args.upload_id, meta=meta)
except KeyError as err:
api.patch("contentupload", args.upload_id, status="Error: malformed AmpliSeq archive")
api.post('log', upload='/rundb/api/v1/contentupload/%s/' % str(args.upload_id), text="Malformed AmpliSeq archive: missing json key "+str(err))
raise
if target_regions_bed and target_regions_bed not in files:
api.patch("contentupload", args.upload_id, status="Error: malformed AmpliSeq archive")
api.post('log', upload='/rundb/api/v1/contentupload/%s/' % str(args.upload_id),
text="Malformed AmpliSeq archive: Target region file %s not present in AmpliSeq archive" % target_regions_bed)
raise ValueError("Target region file %s not present in AmpliSeq archive" % target_regions_bed)
if hotspots_bed and hotspots_bed not in files:
api.patch("contentupload", args.upload_id, status="Error: malformed AmpliSeq archive")
api.post('log', upload='/rundb/api/v1/contentupload/%s/' % str(args.upload_id),
text="Malformed AmpliSeq archive: Hotspots file %s not present in AmpliSeq archive" % target_regions_bed)
raise ValueError("Hotspots file %s not present in AmpliSeq archive" % target_regions_bed)
else:
api.patch("contentupload", args.upload_id, status="Error: Upload must be either valid Ampliseq export or contain a single BED or VCF file.")
raise ValueError("Upload must be either valid Ampliseq export or contain a single BED or VCF file.")
''' === Validate and Register === '''
primary_path = None
secondary_path = None
if target_regions_bed:
primary_path = validate(args.upload_id, args.path, meta, target_regions_bed, 'target regions BED')
if hotspots_bed:
secondary_path = validate(args.upload_id, args.path, meta, hotspots_bed, 'hotspots BED')
meta["hotspot"] = False
if target_regions_bed and not primary_path:
register_bed_file(args.upload_id, args.path, meta, target_regions_bed)
if hotspots_bed:
meta["hotspot"] = True
if not secondary_path:
register_bed_file(args.upload_id, args.path, meta, hotspots_bed)
if meta['is_ampliseq']:
try:
if target_regions_bed and not primary_path:
primary_path = os.path.join(args.path, meta["reference"]+"/unmerged/detail/"+target_regions_bed)
if hotspots_bed and not secondary_path:
secondary_path = os.path.join(args.path, meta["reference"]+"/unmerged/detail/"+hotspots_bed)
plan_prototype = plan_json(meta, primary_path, secondary_path)
api.post("plannedexperiment", **plan_prototype)
except Exception as err:
print("Could not create plan from this zip: %s" % err)
meta_file_handle = open(args.meta_file,'w')
json.dump(meta, meta_file_handle, cls=JSONEncoder)
meta_file_handle.close()
api.patch("contentupload", args.upload_id, meta=meta)
def main():
parse = argparse.ArgumentParser()
parse.add_argument('upload_id', type=int)
parse.add_argument('path')
parse.add_argument('upload_file')
parse.add_argument('meta_file')
try:
args = parse.parse_args()
except IOError as err:
print("ERROR: Input file error: %s" % err)
parse.print_help()
sys.exit(1)
meta_file_handle = open(args.meta_file, 'r')
meta = json.load(meta_file_handle, parse_float=Decimal)
meta_file_handle.close()
print "Uploaded file: " + os.path.basename(args.upload_file)
is_zip = zipfile.is_zipfile(args.upload_file)
if is_zip:
files = file_utils.unzip_archive(args.path, args.upload_file)
print "Compressed: Yes (zip)"
elif args.upload_file.endswith('.gz'):
print "Compressed: Yes (gzip)"
files = [os.path.basename(args.upload_file[:-3])]
cmd = 'gzip -dc %s > %s ' % (args.upload_file,
os.path.join(args.path, files[0]))
p = subprocess.Popen(cmd,
stderr=subprocess.STDOUT,
stdout=subprocess.PIPE,
shell=True)
print p.communicate()[0]
if p.returncode != 0:
sys.exit(p.returncode)
subprocess.call(cmd, shell=True)
else:
print "Compressed: No"
files = [args.upload_file]
''' Establish the upload type '''
target_regions_bed = None
hotspots_bed = None
meta['is_ampliseq'] = False
if len(files) == 1 and files[0].endswith('.bed') and meta.get(
'hotspot', False) == False:
target_regions_bed = os.path.basename(files[0])
meta['is_ampliseq'] = False
print "Content: Target regions file in BED format"
print
elif len(files) == 1 and files[0].endswith('.bed') and meta.get(
'hotspot', False) == True:
hotspots_bed = os.path.basename(files[0])
meta['is_ampliseq'] = False
print "Content: Hotspots file in BED format"
print
elif len(files) == 1 and files[0].endswith('.vcf') and meta.get(
'hotspot', False) == True:
print "Content: Hotspots file in VCF format"
print
print "Converting hotspot VCF file to BED: %s" % files[0]
print
hotspots_bed = os.path.basename(files[0]) + '.bed'
convert_command = '/usr/local/bin/tvcutils prepare_hotspots'
convert_command += ' --input-vcf %s' % os.path.join(
args.path, os.path.basename(files[0]))
convert_command += ' --output-bed %s' % os.path.join(
args.path, hotspots_bed)
convert_command += ' --reference /results/referenceLibrary/tmap-f3/%s/%s.fasta' % (
meta["reference"], meta["reference"])
convert_command += ' --filter-bypass on'
p = subprocess.Popen(convert_command,
stderr=subprocess.STDOUT,
stdout=subprocess.PIPE,
shell=True)
print p.communicate()[0]
if p.returncode != 0:
sys.exit(p.returncode)
#process = subprocess.Popen(convert_command, stderr=subprocess.STDOUT, stdout=subprocess.PIPE, shell=True)
#for line in process.communicate()[0].splitlines():
# api.post('log', upload='/rundb/api/v1/contentupload/%s/' % str(args.upload_id), text=line.strip())
meta['is_ampliseq'] = False
elif "plan.json" in files:
print "Content: AmpliSeq ZIP"
print
meta['is_ampliseq'] = True
plan_data = json.load(open(os.path.join(args.path, "plan.json")))
version, design, meta = ampliseq.handle_versioned_plans(
plan_data, meta)
meta['design'] = design
try:
target_regions_bed = design['plan']['designed_bed']
hotspots_bed = design['plan']['hotspot_bed']
if not meta.get("reference", None):
meta['reference'] = design['genome'].lower()
if 'design_name' in plan_data:
meta['description'] = design['design_name']
meta_file_handle = open(args.meta_file, 'w')
json.dump(meta, meta_file_handle, cls=JSONEncoder)
meta_file_handle.close()
api.patch("contentupload", args.upload_id, meta=meta)
except KeyError as err:
api.patch("contentupload",
args.upload_id,
status="Error: malformed AmpliSeq archive")
print "ERROR: Malformed AmpliSeq archive: missing json key " + str(
err)
sys.exit(1)
#api.post('log', upload='/rundb/api/v1/contentupload/%s/' % str(args.upload_id), text="Malformed AmpliSeq archive: missing json key "+str(err))
#raise
if target_regions_bed and target_regions_bed not in files:
api.patch("contentupload",
args.upload_id,
status="Error: malformed AmpliSeq archive")
print "ERROR: Target region file %s not present in AmpliSeq archive" % target_regions_bed
sys.exit(1)
#api.post('log', upload='/rundb/api/v1/contentupload/%s/' % str(args.upload_id),
# text="Malformed AmpliSeq archive: Target region file %s not present in AmpliSeq archive" % target_regions_bed)
#raise ValueError("Target region file %s not present in AmpliSeq archive" % target_regions_bed)
if hotspots_bed and hotspots_bed not in files:
api.patch("contentupload",
args.upload_id,
status="Error: malformed AmpliSeq archive")
print "ERROR: Hotspots file %s not present in AmpliSeq archive" % target_regions_bed
sys.exit(1)
#api.post('log', upload='/rundb/api/v1/contentupload/%s/' % str(args.upload_id),
# text="Malformed AmpliSeq archive: Hotspots file %s not present in AmpliSeq archive" % target_regions_bed)
#raise ValueError("Hotspots file %s not present in AmpliSeq archive" % target_regions_bed)
else:
api.patch("contentupload",
args.upload_id,
status="Error: Unrecognized upload type.")
print
print "ERROR: Unrecognized upload type. Upload must be either a valid Ampliseq ZIP or contain a single BED or VCF file."
sys.exit(1)
''' === Validate and Register === '''
primary_path = None
secondary_path = None
if target_regions_bed:
primary_path = validate(args.upload_id, args.path, meta,
target_regions_bed, 'target regions BED')
if hotspots_bed:
secondary_path = validate(args.upload_id, args.path, meta,
hotspots_bed, 'hotspots BED')
meta["hotspot"] = False
if target_regions_bed and not primary_path:
register_bed_file(args.upload_id, args.path, meta, target_regions_bed)
if hotspots_bed:
meta["hotspot"] = True
if not secondary_path:
register_bed_file(args.upload_id, args.path, meta, hotspots_bed)
if meta['is_ampliseq']:
try:
if target_regions_bed and not primary_path:
primary_path = os.path.join(
args.path, meta["reference"] + "/unmerged/detail/" +
target_regions_bed)
if hotspots_bed and not secondary_path:
secondary_path = os.path.join(
args.path,
meta["reference"] + "/unmerged/detail/" + hotspots_bed)
plan_prototype = plan_json(meta, args.upload_id, primary_path,
secondary_path)
success, response, content = api.post("plannedexperiment",
**plan_prototype)
if not success:
api.patch("contentupload",
args.upload_id,
status="Error: unable to create TS Plan")
raise Exception("Plan creation API request failed.")
except Exception as err:
print("ERROR: Could not create plan from this zip: %s" % err)
raise
meta_file_handle = open(args.meta_file, 'w')
json.dump(meta, meta_file_handle, cls=JSONEncoder)
meta_file_handle.close()
api.patch("contentupload", args.upload_id, meta=meta)
def pre_process():
parse = argparse.ArgumentParser()
parse.add_argument('upload_id', type=int)
parse.add_argument('path')
parse.add_argument('upload_file')
parse.add_argument('meta_file', type=argparse.FileType('r+'))
try:
args = parse.parse_args()
except IOError as err:
print("Input file error: %s" % err)
parse.print_help()
sys.exit(1)
meta = json.load(args.meta_file, parse_float=Decimal)
meta.update({
"is_ampliseq": None,
"primary_bed": None,
"hotspot_bed": None
})
is_zip = zipfile.is_zipfile(args.upload_file)
if is_zip:
files = unzip_archive(args.path, args.upload_file)
else:
files = [args.upload_file]
if len(files) == 1 and files[0].endswith('.bed'):
meta['is_ampliseq'] = False
meta['primary_bed'] = files[0]
elif len(files) == 1 and files[0].endswith('.vcf') and meta['hotspot']:
# convert vcf to bed
target_filename = os.path.join(args.path, os.path.basename(
files[0])) + '.bed'
convert_command = '/usr/local/bin/tvcutils prepare_hotspots'
convert_command += ' --input-vcf %s' % files[0]
convert_command += ' --output-bed %s' % target_filename
convert_command += ' --reference /results/referenceLibrary/tmap-f3/%s/%s.fasta' % (
meta["reference"], meta["reference"]
) #TODO: is this just name or full path??
convert_command += ' --filter-bypass on'
process = subprocess.Popen(convert_command,
stderr=subprocess.STDOUT,
stdout=subprocess.PIPE,
shell=True)
for line in process.communicate()[0].splitlines():
api.post('log',
upload='/rundb/api/v1/contentupload/%s/' %
str(args.upload_id),
text=line.strip())
meta['is_ampliseq'] = False
meta['primary_bed'] = target_filename
'''
elif len(files) == 1 and files[0].endswith('.vcf') and meta['hotspot']:
# convert vcf to bed
target_filename = os.path.join(args.path,os.path.basename(files[0])) + '.bed'
convert_command = '/usr/local/bin/tvcutils prepare_hotspots'
convert_command += ' --input-vcf %s' % files[0]
convert_command += ' --output-bed %s' % target_filename
convert_command += ' --reference /results/referenceLibrary/tmap-f3/%s/%s.fasta' % (meta["reference"],meta["reference"]) #TODO: is this just name or full path??
convert_command += ' --filter-bypass on'
process = subprocess.Popen(convert_command, stderr=subprocess.STDOUT, stdout=subprocess.PIPE, shell=True)
for line in process.communicate()[0]:
api.post('log', upload='/rundb/api/v1/contentupload/%s/' % str(args.upload_id), text=line.strip())
meta['is_ampliseq'] = False
meta['primary_bed'] = target_filename
'''
elif "plan.json" in files:
print("Found ampliseq")
meta['is_ampliseq'] = True
plan_data = json.load(open(os.path.join(args.path, "plan.json")),
parse_float=Decimal)
version, design = ampliseq.handle_versioned_plans(plan_data)
meta['design'] = design
plan = design['plan']
try:
meta['primary_bed'] = plan['designed_bed']
meta['secondary_bed'] = plan['hotspot_bed']
if not meta.get("reference", None):
meta['reference'] = design['genome'].lower()
except KeyError as err:
api.patch("contentupload",
args.upload_id,
status="Error: malformed AmpliSeq archive")
raise
print(meta)
else:
raise ValueError(
"Upload must be either valid Ampliseq export or contain a single BED file."
)
args.meta_file.truncate(0)
args.meta_file.seek(0)
json.dump(meta, args.meta_file, cls=JSONEncoder)
api.patch("contentupload", args.upload_id, meta=meta)