def upload_STARalignment(self, input_params, reads_ref, reads_info, output_bam_file):
"""
Uploads the alignment file + metadata.
Returns the STAR alignment reference.
"""
aligner_opts = dict()
for k in input_params:
aligner_opts[k] = str(input_params[k])
pprint(reads_info)
alignment_name = reads_ref['alignment_output_name']
align_upload_params = {
"destination_ref": "{}/{}".format(input_params[self.PARAM_IN_WS], alignment_name),
"file_path": output_bam_file,
"assembly_or_genome_ref": input_params[self.PARAM_IN_GENOME],
"read_library_ref": reads_info['object_ref'],
"library_type": reads_info['style'],
"condition": reads_info['condition'],
"aligned_using": 'STAR',
"aligner_version":self.STAR_VERSION,
"aligner_opts": aligner_opts
}
pprint(align_upload_params)
ra_util = ReadsAlignmentUtils(self.callback_url, service_ver='beta')
rau_upload_ret = ra_util.upload_alignment(align_upload_params)
alignment_ref = rau_upload_ret["obj_ref"]
print("STAR alignment uploaded as object {}".format(alignment_ref))
return rau_upload_ret
def _gen_ctab_files(self, params, alignment_ref):
source_dir = params.get(self.PARAM_IN_SRC_DIR)
if len(glob.glob(source_dir + '/*.ctab')) < 5:
self.__LOGGER.info(' ======= Generating ctab files ==========')
gtf_file = os.path.join(source_dir, 'transcripts.gtf')
if not os.path.isfile(gtf_file):
raise ValueError("{} file is required to generate ctab files, found missing".
format(gtf_file))
if self.PARAM_IN_BAM_FILE_PATH in params and \
params[self.PARAM_IN_BAM_FILE_PATH] is not None:
bam_file_path = params[self.PARAM_IN_BAM_FILE_PATH]
else:
self.__LOGGER.info('Downloading bam file from alignment object')
rau = ReadsAlignmentUtils(self.callback_url)
alignment_retVal = rau.download_alignment({'source_ref': alignment_ref})
alignment_dir = alignment_retVal.get('destination_dir')
tmp_file_path = os.path.join(alignment_dir, 'accepted_hits.bam')
if os.path.isfile(tmp_file_path):
bam_file_path = tmp_file_path
else:
tmp_file_path = os.path.join(alignment_dir, 'accepted_hits_sorted.bam')
if os.path.isfile(tmp_file_path):
bam_file_path = tmp_file_path
else:
raise ValueError('accepted_hits.bam or accepted_hits_sorted.bam not found in {}'.
format(alignment_dir))
result = self.table_maker.build_ctab_files(
ref_genome_path=gtf_file,
alignment_path=bam_file_path,
output_dir=source_dir)
if result != 0:
raise ValueError('Tablemaker failed')
def __init__(self, config):
"""
:param config:
:param logger:
:param directory: Working directory
:param urls: Service urls
"""
# BEGIN_CONSTRUCTOR
self.ws_url = config["workspace-url"]
self.ws_url = config["workspace-url"]
self.callback_url = config['SDK_CALLBACK_URL']
self.srv_wiz_url = config['srv-wiz-url']
self.token = config['KB_AUTH_TOKEN']
self.shock_url = config['shock-url']
self.dfu = DataFileUtil(self.callback_url)
self.gfu = GenomeFileUtil(self.callback_url)
self.au = AssemblyUtil(self.callback_url)
self.rau = ReadsAlignmentUtils(self.callback_url)
self.set_api = SetAPI(self.srv_wiz_url, service_ver='dev')
self.eu = ExpressionUtils(self.callback_url)
self.ws = Workspace(self.ws_url, token=self.token)
self.scratch = os.path.join(config['scratch'], str(uuid.uuid4()))
self._mkdir_p(self.scratch)
self.tool_used = "Cufflinks"
self.tool_version = os.environ['VERSION']
# END_CONSTRUCTOR
pass
def __init__(self, scratch_dir, callback_url, workspace_url, srv_wiz_url):
self.scratch_dir = scratch_dir
self.rau = ReadsAlignmentUtils(callback_url)
self.kbr = KBaseReport(callback_url)
self.dfu = DataFileUtil(callback_url)
self.set_api = SetAPI(srv_wiz_url)
self.ws = Workspace(workspace_url)
self.valid_commands = ['bamqc', 'multi-bamqc']
def __init__(self, config):
log('--->\nInitializing RNASeqDownloaderUtils instance:\n config: %s' %
config)
self.scratch = config['scratch']
self.callback_url = config['SDK_CALLBACK_URL']
self.token = config['KB_AUTH_TOKEN']
self.dfu = DataFileUtil(self.callback_url, token=self.token)
self.rau = ReadsAlignmentUtils(self.callback_url, token=self.token)
def __init__(self, ctx, config):
log('--->\nInitializing RNASeqDownloaderUtils instance:\n config: %s' % config)
self.scratch = config['scratch']
self.callback_url = config['SDK_CALLBACK_URL']
self.token = config['KB_AUTH_TOKEN']
self.ctx = ctx
self.staging_base = os.path.join(self.STAGING_FILE_PREFIX, ctx['user_id'])
self.rau = ReadsAlignmentUtils(self.callback_url, token=self.token)
def save_read_alignment_output(self, run_output_info, input_configuration, validated_params):
rau = ReadsAlignmentUtils(self.callback_url)
destination_ref = validated_params['output_workspace'] + '/' + validated_params['output_alignment_name']
condition = 'unknown'
if 'condition_label' in validated_params:
condition = validated_params['condition_label']
upload_params = {'file_path': run_output_info['output_sam_file'],
'destination_ref': destination_ref,
'read_library_ref': input_configuration['reads_lib_ref'],
'assembly_or_genome_ref': validated_params['assembly_or_genome_ref'],
'condition': condition}
upload_results = rau.upload_alignment(upload_params)
return upload_results
def load_bam_file(self, file_path, genome_ref, reads_ref, obj_name):
rau = ReadsAlignmentUtils(self.callback_url, service_ver='dev')
return rau.upload_alignment({
"destination_ref":
"{}/{}".format(self.ws_name, obj_name),
"file_path":
file_path,
"read_library_ref":
reads_ref,
"assembly_or_genome_ref":
genome_ref,
"condition":
"none"
})["obj_ref"]
def setUpClass(cls):
token = environ.get('KB_AUTH_TOKEN', None)
config_file = environ.get('KB_DEPLOYMENT_CONFIG', None)
cls.cfg = {}
config = ConfigParser()
config.read(config_file)
for nameval in config.items('AlignmentSetEditor'):
cls.cfg[nameval[0]] = nameval[1]
# Getting username from Auth profile for token
authServiceUrl = cls.cfg['auth-service-url']
auth_client = _KBaseAuth(authServiceUrl)
user_id = auth_client.get_user(token)
# WARNING: don't call any logging methods on the context object,
# it'll result in a NoneType error
cls.ctx = MethodContext(None)
cls.ctx.update({'token': token,
'user_id': user_id,
'provenance': [
{'service': 'AlignmentSetEditor',
'method': 'please_never_use_it_in_production',
'method_params': []
}],
'authenticated': 1})
cls.wsURL = cls.cfg['workspace-url']
cls.wsClient = workspaceService(cls.wsURL)
cls.serviceImpl = AlignmentSetEditor(cls.cfg)
cls.scratch = cls.cfg['scratch']
cls.callback_url = os.environ['SDK_CALLBACK_URL']
cls.setAPI = SetAPI(cls.callback_url)
cls.gfu = GenomeFileUtil(cls.callback_url)
cls.ru = ReadsUtils(cls.callback_url)
cls.rau = ReadsAlignmentUtils(cls.callback_url)
suffix = int(time.time() * 1000)
cls.wsName = "test_AlignmentSetEditor_" + str(suffix)
cls.wsClient.create_workspace({'workspace': cls.wsName})
def setUpClass(cls):
token = environ.get('KB_AUTH_TOKEN', None)
config_file = environ.get('KB_DEPLOYMENT_CONFIG', None)
cls.cfg = {}
config = ConfigParser()
config.read(config_file)
for nameval in config.items('kb_cufflinks'):
cls.cfg[nameval[0]] = nameval[1]
# Getting username from Auth profile for token
authServiceUrl = cls.cfg['auth-service-url']
auth_client = _KBaseAuth(authServiceUrl)
user_id = auth_client.get_user(token)
# WARNING: don't call any logging methods on the context object,
# it'll result in a NoneType error
cls.ctx = MethodContext(None)
cls.ctx.update({'token': token,
'user_id': user_id,
'provenance': [
{'service': 'kb_cufflinks',
'method': 'please_never_use_it_in_production',
'method_params': []
}],
'authenticated': 1})
cls.wsURL = cls.cfg['workspace-url']
cls.wsClient = Workspace(url=cls.wsURL, token=token)
cls.serviceImpl = kb_cufflinks(cls.cfg)
cls.scratch = cls.cfg['scratch']
cls.callback_url = environ.get('SDK_CALLBACK_URL')
cls.srv_wiz_url = cls.cfg['srv-wiz-url']
# cls.wsName = 'cufflinks_test_' + user_id # reuse existing workspace
suffix = int(time.time() * 1000)
cls.wsName = "test_kb_cufflinks_" + str(suffix)
print('workspace_name: ' + cls.wsName)
try:
# reuse existing (previously torn down) workspace
cls.wsClient.undelete_workspace({'workspace': cls.wsName})
print('reusing old workspace...')
except BaseException:
try:
# create if workspace does not exist
cls.wsClient.create_workspace({'workspace': cls.wsName})
except BaseException:
# get workspace if it exists and was not previously deleted (previously
# not torn down)
ws_info = cls.wsClient.get_workspace_info({'workspace': cls.wsName})
print("creating new workspace: " + str(ws_info))
cls.dfu = DataFileUtil(cls.callback_url)
cls.gfu = GenomeFileUtil(cls.callback_url)
cls.ru = ReadsUtils(cls.callback_url)
cls.rau = ReadsAlignmentUtils(cls.callback_url)
cls.set_api = SetAPI(cls.srv_wiz_url, service_ver='dev')
cls.cufflinks_runner = CufflinksUtils(cls.cfg)
cls.prepare_data()
def get_alignment_data_files(self, alignment_refs):
"""
Returns a dictionary of data files. Key = object name, value = path to the file.
"""
alignment_files = dict()
ru = ReadsAlignmentUtils(self.callback_url, service_ver='dev')
for ref in alignment_refs:
ref_name = get_object_name(ref, self.workspace_url)
align_file = ru.download_alignment({
"source_ref": ref,
"downloadBAI": 0
})
for f in os.listdir(align_file["destination_dir"]):
if f.endswith("bam"):
alignment_files[ref_name] = os.path.join(
align_file["destination_dir"], f)
return alignment_files
def setUpClass(cls):
cls.token = environ.get('KB_AUTH_TOKEN', None)
cls.callbackURL = environ.get('SDK_CALLBACK_URL')
config_file = environ.get('KB_DEPLOYMENT_CONFIG', None)
cls.cfg = {}
config = ConfigParser()
config.read(config_file)
for nameval in config.items('ExpressionUtils'):
cls.cfg[nameval[0]] = nameval[1]
# Getting username from Auth profile for token
authServiceUrl = cls.cfg['auth-service-url']
auth_client = _KBaseAuth(authServiceUrl)
user_id = auth_client.get_user(cls.token)
# WARNING: don't call any logging methods on the context object,
# it'll result in a NoneType error
cls.ctx = MethodContext(None)
cls.ctx.update({
'token':
cls.token,
'user_id':
user_id,
'provenance': [{
'service': 'ExpressionUtils',
'method': 'please_never_use_it_in_production',
'method_params': []
}],
'authenticated':
1
})
cls.shockURL = cls.cfg['shock-url']
cls.wsURL = cls.cfg['workspace-url']
cls.service_wizard_url = cls.cfg['srv-wiz-url']
cls.wsClient = workspaceService(cls.wsURL)
cls.ws = Workspace(cls.wsURL, token=cls.token)
cls.hs = HandleService(url=cls.cfg['handle-service-url'],
token=cls.token)
# create workspace
wssuffix = int(time.time() * 1000)
wsname = "test_expression_" + str(wssuffix)
cls.wsinfo = cls.wsClient.create_workspace({'workspace': wsname})
print('created workspace ' + cls.getWsName())
cls.serviceImpl = ExpressionUtils(cls.cfg)
cls.readUtils = ReadsUtils(cls.callbackURL)
cls.dfu = DataFileUtil(cls.callbackURL, service_ver='dev')
cls.dfu.ws_name_to_id(wsname)
cls.assemblyUtil = AssemblyUtil(cls.callbackURL)
cls.gfu = GenomeFileUtil(cls.callbackURL)
cls.gaAPI = GenomeAnnotationAPI(cls.service_wizard_url)
cls.rau = ReadsAlignmentUtils(cls.callbackURL)
cls.scratch = cls.cfg['scratch']
cls.staged = {}
cls.nodes_to_delete = []
cls.handles_to_delete = []
cls.setupTestData()
def __init__(self, config):
self.ws_url = config["workspace-url"]
self.callback_url = config['SDK_CALLBACK_URL']
self.token = config['KB_AUTH_TOKEN']
self.scratch = config['scratch']
self.dfu = DataFileUtil(self.callback_url)
self.ru = ReadsUtils(self.callback_url)
self.au = AssemblyUtil(self.callback_url)
self.gfu = GenomeFileUtil(self.callback_url)
self.rau = ReadsAlignmentUtils(self.callback_url)
def upload_alignment(self, input_params, reads_info, alignment_name,
alignment_file):
"""
Uploads the alignment file + metadata.
This then returns the expected return dictionary from HISAT2.
"""
aligner_opts = dict()
for k in input_params:
aligner_opts[k] = str(input_params[k])
align_upload_params = {
"destination_ref":
"{}/{}".format(input_params["ws_name"], alignment_name),
"file_path":
alignment_file,
"library_type":
reads_info["style"], # single or paired end,
"condition":
reads_info["condition"],
"assembly_or_genome_ref":
input_params["genome_ref"],
"read_library_ref":
reads_info["object_ref"],
"aligned_using":
"hisat2",
"aligner_version":
HISAT_VERSION,
"aligner_opts":
aligner_opts
}
if "sampleset_ref" in reads_info:
align_upload_params["sampleset_ref"] = reads_info["sampleset_ref"]
print("Uploading completed alignment")
pprint(align_upload_params)
ra_util = ReadsAlignmentUtils(self.callback_url, service_ver="dev")
alignment_ref = ra_util.upload_alignment(
align_upload_params)["obj_ref"]
print(
"Done! New alignment uploaded as object {}".format(alignment_ref))
return alignment_ref
def __init__(self, config):
self.ws_url = config["workspace-url"]
self.callback_url = config['SDK_CALLBACK_URL']
self.token = config['KB_AUTH_TOKEN']
self.shock_url = config['shock-url']
self.dfu = DataFileUtil(self.callback_url)
self.rau = ReadsAlignmentUtils(self.callback_url)
self.fv = KBaseFeatureValues(self.callback_url)
self.deu = DifferentialExpressionUtils(self.callback_url, service_ver='dev')
self.ws = Workspace(self.ws_url, token=self.token)
self.scratch = config['scratch']
self.config = config
def setUpClass(cls):
cls.token = environ.get('KB_AUTH_TOKEN', None)
config_file = environ.get('KB_DEPLOYMENT_CONFIG', None)
cls.cfg = {}
config = ConfigParser()
config.read(config_file)
for nameval in config.items('kb_ballgown'):
cls.cfg[nameval[0]] = nameval[1]
# Getting username from Auth profile for token
authServiceUrl = cls.cfg['auth-service-url']
auth_client = _KBaseAuth(authServiceUrl)
user_id = auth_client.get_user(cls.token)
# WARNING: don't call any logging methods on the context object,
# it'll result in a NoneType error
cls.ctx = MethodContext(None)
cls.ctx.update({
'token':
cls.token,
'user_id':
user_id,
'provenance': [{
'service': 'kb_ballgown',
'method': 'please_never_use_it_in_production',
'method_params': []
}],
'authenticated':
1
})
cls.hs = HandleService(url=cls.cfg['handle-service-url'],
token=cls.token)
cls.shockURL = cls.cfg['shock-url']
cls.wsURL = cls.cfg['workspace-url']
cls.wsClient = workspaceService(cls.wsURL, token=cls.token)
cls.serviceImpl = kb_ballgown(cls.cfg)
cls.scratch = cls.cfg['scratch']
cls.callback_url = os.environ['SDK_CALLBACK_URL']
cls.gfu = GenomeFileUtil(cls.callback_url)
cls.dfu = DataFileUtil(cls.callback_url)
cls.ru = ReadsUtils(cls.callback_url)
cls.rau = ReadsAlignmentUtils(cls.callback_url, service_ver='dev')
cls.eu = ExpressionUtils(cls.callback_url, service_ver='dev')
cls.set_api = SetAPI(cls.callback_url)
suffix = int(time.time() * 1000)
cls.wsName = "test_kb_ballgown_" + str(suffix)
#cls.wsName = "test_kb_ballgown_1004"
cls.wsClient.create_workspace({'workspace': cls.wsName})
cls.nodes_to_delete = []
cls.handles_to_delete = []
cls.prepare_data()
def __init__(self, config):
self.ws_url = config["workspace-url"]
self.callback_url = config['SDK_CALLBACK_URL']
self.token = config['KB_AUTH_TOKEN']
self.shock_url = config['shock-url']
self.srv_wiz_url = config['srv-wiz-url']
self.scratch = config['scratch']
self.dfu = DataFileUtil(self.callback_url)
self.gfu = GenomeFileUtil(self.callback_url)
self.rau = ReadsAlignmentUtils(self.callback_url)
self.au = AssemblyUtil(self.callback_url)
self.eu = ExpressionUtils(self.callback_url, service_ver='dev')
self.ws = Workspace(self.ws_url, token=self.token)
self.set_client = SetAPI(self.srv_wiz_url)
def __init__(self, config):
self.ws_url = config["workspace-url"]
self.callback_url = config['SDK_CALLBACK_URL']
self.token = config['KB_AUTH_TOKEN']
self.shock_url = config['shock-url']
self.scratch = config['scratch']
self.srv_wiz_url = config['srv-wiz-url']
self.ws = Workspace(self.ws_url, token=self.token)
self.bt = kb_Bowtie2(self.callback_url)
self.rau = ReadsAlignmentUtils(self.callback_url)
self.qualimap = kb_QualiMap(self.callback_url)
self.ru = ReadsUtils(self.callback_url)
self.dfu = DataFileUtil(self.callback_url)
self.set_client = SetAPI(self.srv_wiz_url)
def upload_alignment(self, file_path, reads_upa, assembly_upa, workspace_name, alignment_name):
if not file_path:
raise ValueError("file_path must be defined")
if not os.path.exists(file_path):
raise ValueError("The given alignment file '{}' does not exist".format(file_path))
if not reads_upa:
raise ValueError("The reads UPA must be defined")
if not assembly_upa:
raise ValueError("The assembly UPA must be defined")
if not workspace_name:
raise ValueError("workspace_name must be defined")
if not alignment_name:
raise ValueError("alignment_name must be defined")
rau = ReadsAlignmentUtils(self.callback_url)
alignment_upa = rau.upload_alignment({
"file_path": file_path,
"read_library_ref": reads_upa,
"assembly_or_genome_ref": assembly_upa,
"destination_ref": "{}/{}".format(workspace_name, alignment_name),
"aligned_using": "BBMap",
"condition": "new_assembly"
})["obj_ref"]
return alignment_upa
def __init__(self, config, services, logger=None):
self.config = config
self.logger = logger
self.callback_url = os.environ['SDK_CALLBACK_URL']
self.scratch = os.path.join(config['scratch'],
'cuffdiff_merge_' + str(uuid.uuid4()))
self.ws_url = config['workspace-url']
self.services = services
self.ws_client = Workspace(self.services['workspace_service_url'])
self.dfu = DataFileUtil(self.callback_url)
self.gfu = GenomeFileUtil(self.callback_url)
self.rau = ReadsAlignmentUtils(self.callback_url)
self.eu = ExpressionUtils(self.callback_url)
self.deu = DifferentialExpressionUtils(self.callback_url)
self.cuffmerge_runner = CuffMerge(config, logger)
self.num_threads = mp.cpu_count()
handler_utils._mkdir_p(self.scratch)
def setUpClass(cls):
token = environ.get('KB_AUTH_TOKEN', None)
cls.callback_url = os.environ['SDK_CALLBACK_URL']
cls.ctx = {'token': token}
config_file = environ.get('KB_DEPLOYMENT_CONFIG', None)
cls.cfg = {}
config = ConfigParser()
config.read(config_file)
for nameval in config.items('kb_cummerbund'):
cls.cfg[nameval[0]] = nameval[1]
cls.scratch = cls.cfg['scratch']
cls.wsURL = cls.cfg['ws_url']
cls.gfu = GenomeFileUtil(cls.callback_url)
cls.ru = ReadsUtils(cls.callback_url)
cls.rau = ReadsAlignmentUtils(cls.callback_url, service_ver='dev')
cls.eu = ExpressionUtils(cls.callback_url, service_ver='dev')
cls.wsClient = workspaceService(cls.wsURL, token=token)
cls.dfu = DataFileUtil(cls.callback_url, token=token)
cls.serviceImpl = kb_cummerbund(cls.cfg)
cls.prepare_data()
class RNASeqExportUtils:
STAGING_FILE_PREFIX = '/staging/'
def __init__(self, ctx, config):
log('--->\nInitializing RNASeqDownloaderUtils instance:\n config: %s' % config)
self.scratch = config['scratch']
self.callback_url = config['SDK_CALLBACK_URL']
self.token = config['KB_AUTH_TOKEN']
self.ctx = ctx
self.staging_base = os.path.join(self.STAGING_FILE_PREFIX, ctx['user_id'])
self.rau = ReadsAlignmentUtils(self.callback_url, token=self.token)
def download_RNASeq_Alignment_BAM(self, params):
"""
download_RNASeq: download RNASeq Alignment/Expression/DifferentialExpression zip file
params:
input_ref: RNASeq object reference ID
rna_seq_type: 'RNASeqAlignment'
download_file_type: one of 'bam', 'sam' or 'bai'
return:
shock_id: Shock ID of stored zip file
"""
log('--->\nrunning RNASeqDownloaderUtils.download_RNASeq_Alignment:\nparams: %s' % params)
# Validate params
self.validate_download_rna_seq_alignment_parameters(params)
input_ref = params.get('input_ref')
returnVal = dict()
tmp_dir = self.rau.download_alignment({'source_ref': input_ref,
'downloadBAI': False})['destination_dir']
files = os.listdir(tmp_dir)
destination_dir = os.path.join(self.staging_base, params['destination_dir'])
if not os.path.exists(destination_dir):
os.mkdir(destination_dir)
for fn in files:
print self.staging_base
shutil.move(os.path.join(tmp_dir, fn), destination_dir)
returnVal['path'] = destination_dir
return returnVal
def download_RNASeq_Alignment_SAM(self, params):
"""
download_RNASeq: download RNASeq Alignment/Expression/DifferentialExpression zip file
params:
input_ref: RNASeq object reference ID
rna_seq_type: 'RNASeqAlignment'
return:
"""
log('--->\nrunning RNASeqDownloaderUtils.download_RNASeq_Alignment:\nparams: %s' % params)
destination_dir = self.rau.download_alignment({'source_ref': input_ref,
'downloadSAM': True,
'downloadBAI': True})['destination_dir']
files = os.listdir(destination_dir)
bam_files = [x for x in files if re.match('.*\.bam', x)]
for bam_file in bam_files:
log('removing file: {}'.format(bam_file))
os.remove(os.path.join(destination_dir, bam_file))
files = os.listdir(destination_dir)
for fn in files:
shutil.move(os.path.join(destination_dir, fn),
params['destination_dir'])
return {'path': params['destination_dir']}
def validate_download_rna_seq_alignment_parameters(self, params):
"""
validate_download_rna_seq_alignment_parameters:
validates params passed to download_rna_seq_alignment method
"""
# check required parameters
for p in ['input_ref', 'destination_dir']:
if p not in params:
raise ValueError('"' + p + '" parameter is required, but missing')
# check supportive RNASeq types
valid_rnaseq_types = ['RNASeqAlignment',
'RNASeqExpression',
'RNASeqDifferentialExpression']
if params['rna_seq_type'] not in valid_rnaseq_types:
raise ValueError('Unexpected RNASeq type: %s' % params['rna_seq_type'])
class QualiMapRunner:
QUALIMAP_PATH = '/kb/module/qualimap-bin/qualimap'
JAVA_MEM_DEFAULT_SIZE = '16G'
LARGE_BAM_FILE_SIZE = 20 * 1024 * 1024 * 1024 # 20 GB
TIMEOUT = 72 * 60 * 60 # 72 hours
def _get_file_size(self, file_path):
file_size = os.path.getsize(file_path)
print('File size: {} -- {}'.format(file_size, file_path))
return file_size
def _large_file(self, file_path):
filename, file_extension = os.path.splitext(file_path)
multiplier = 0
if file_extension == '.txt':
total_file_size = 0
with open(file_path, 'r') as f:
for line in f:
bam_file_path = line.split('\t')[1]
total_file_size += self._get_file_size(bam_file_path)
print('Total file size: {}'.format(total_file_size))
multiplier = int(total_file_size) / int(self.LARGE_BAM_FILE_SIZE)
else:
multiplier = int(self._get_file_size(file_path)) / int(
self.LARGE_BAM_FILE_SIZE)
print('setting number of windows multiplier to: {}'.format(multiplier))
return multiplier
def _timeout_handler(self, signum, frame):
print('Signal handler called with signal', signum)
raise ValueError('QualiMap takes too long')
def __init__(self, scratch_dir, callback_url, workspace_url, srv_wiz_url):
self.scratch_dir = scratch_dir
self.rau = ReadsAlignmentUtils(callback_url)
self.kbr = KBaseReport(callback_url)
self.dfu = DataFileUtil(callback_url)
self.gfu = GenomeFileUtil(callback_url)
self.set_api = SetAPI(srv_wiz_url)
self.ws = Workspace(workspace_url)
self.valid_commands = ['bamqc', 'multi-bamqc']
def run_app(self, params):
self.validate_params(params)
print('Validated Params = ')
pprint(params)
run_info = self.get_run_info(params)
if run_info.get('mode') not in ['single', 'multi']:
raise ValueError(
'Error in fetching the type to determine run settings.')
run_error = False
try:
signal.signal(signal.SIGALRM, self._timeout_handler)
signal.alarm(self.TIMEOUT)
if run_info['mode'] == 'single':
result = self.run_bamqc(params['input_ref'],
run_info['input_info'])
elif run_info['mode'] == 'multi':
result = self.run_multi_sample_qc(params['input_ref'],
run_info['input_info'])
signal.alarm(0)
except Exception:
run_error = True
workdir = os.path.join(self.scratch_dir,
'qualimap_' + str(int(time.time() * 10000)))
os.makedirs(workdir)
with open(os.path.join(workdir, 'qualimapReport.html'),
'w') as report:
report.write('<html><body><p></p></body></html>')
package_info = self.package_output_folder(
workdir, 'QualiMap_report',
'EMPTY HTML report directory for QualiMap BAM QC',
'qualimapReport.html')
result = {
'qc_result_folder_path': workdir,
'qc_result_zip_info': package_info,
'shock_id': None
}
error_msg = 'Running QualiMap returned an error:\n{}\n'.format(
traceback.format_exc())
error_msg += 'Generating simple report instead\n'
print(error_msg)
if params['create_report']:
result = self.create_report(result, params['output_workspace'],
run_error, params['input_ref'])
return result
def create_report(self,
result,
output_workspace,
run_error=None,
input_ref=None):
if run_error:
objects_created = []
info = self.get_obj_info(input_ref)
obj_type = self.get_type_from_obj_info(info)
if obj_type in ['KBaseRNASeq.RNASeqAlignment']:
objects_created.append({
'ref': input_ref,
'description': 'Alignment'
})
if obj_type in [
'KBaseRNASeq.RNASeqAlignmentSet',
'KBaseSets.ReadsAlignmentSet'
]:
objects_created.append({
'ref': input_ref,
'description': 'AlignmentSet'
})
reads_alignment_info = self.get_alignments_from_set(input_ref)
for alignment in reads_alignment_info:
alignment_ref = alignment.get('ref')
objects_created.append({
'ref': alignment_ref,
'description': 'Alignment'
})
report_info = self.kbr.create_extended_report({
'message':
' ',
'objects_created':
objects_created,
'report_object_name':
'qualimap_report' + str(uuid.uuid4()),
'workspace_name':
output_workspace
})
result['report_name'] = report_info['name']
result['report_ref'] = report_info['ref']
return result
qc_result_zip_info = result['qc_result_zip_info']
report_info = self.kbr.create_extended_report({
'message':
'',
'objects_created': [],
'direct_html_link_index':
0,
'html_links': [{
'shock_id': qc_result_zip_info['shock_id'],
'name': qc_result_zip_info['index_html_file_name'],
'label': qc_result_zip_info['name']
}],
'report_object_name':
'qualimap_report' + str(uuid.uuid4()),
'workspace_name':
output_workspace
})
result['report_name'] = report_info['name']
result['report_ref'] = report_info['ref']
return result
def get_gtf_file(self, input_ref, set_op=False):
print('Start fetching GFF file from genome')
if set_op:
set_data = self.set_api.get_reads_alignment_set_v1({
'ref':
input_ref,
'include_item_info':
1
})
input_ref = set_data['data']['items'][0]['ref']
obj_data = self.dfu.get_objects({"object_refs":
[input_ref]})['data'][0]['data']
genome_ref = obj_data.get('genome_id')
if not genome_ref:
raise ValueError(
'Alignment is not associated with a Genome object')
result_directory = os.path.join(self.scratch_dir, str(uuid.uuid4()))
os.makedirs(result_directory)
genome_gtf_file = self.gfu.genome_to_gff({
'genome_ref': genome_ref,
'is_gtf': True,
'target_dir': result_directory
})['file_path']
return genome_gtf_file
def run_bamqc(self, input_ref, input_info):
# download the input and setup a working dir
alignment_info = self.rau.download_alignment({'source_ref': input_ref})
bam_file_path = self.find_my_bam_file(
alignment_info['destination_dir'])
try:
gtf_file = self.get_gtf_file(input_ref)
except:
gtf_file = ''
workdir = os.path.join(self.scratch_dir,
'qualimap_' + str(int(time.time() * 10000)))
options = [
'-bam', bam_file_path, '-c', '-outdir', workdir, '-outformat',
'html'
]
if gtf_file:
options += ['-gff', gtf_file]
options.append('--java-mem-size={}'.format(
self.JAVA_MEM_DEFAULT_SIZE)) # always use large mem
multiplier = self._large_file(bam_file_path)
if multiplier:
window_size = multiplier * 400
print('using larger window size: {} and Java memory: {}'.format(
window_size, self.JAVA_MEM_DEFAULT_SIZE))
options.append(
'-nw {}'.format(window_size)) # increase size of windows
self.run_cli_command('bamqc', options)
package_info = self.package_output_folder(
workdir, 'QualiMap_report',
'HTML report directory for QualiMap BAM QC', 'qualimapReport.html')
return {
'qc_result_folder_path': workdir,
'qc_result_zip_info': package_info
}
def run_multi_sample_qc(self, input_ref, input_info):
# download the input and setup a working dir
reads_alignment_info = self.get_alignments_from_set(input_ref)
try:
gtf_file = self.get_gtf_file(input_ref, set_op=True)
except:
gtf_file = ''
suffix = 'qualimap_' + str(int(time.time() * 10000))
workdir = os.path.join(self.scratch_dir, suffix)
os.makedirs(workdir)
input_file_path = self.create_multi_qualimap_cfg(
reads_alignment_info, workdir)
options = [
'-d', input_file_path, '-r', '-c', '-outdir', workdir,
'-outformat', 'html'
]
if gtf_file:
options += ['-gff', gtf_file]
multiplier = self._large_file(input_file_path)
if multiplier:
window_size = multiplier * 400
print('using larger window size: {} and Java memory: {}'.format(
window_size, self.JAVA_MEM_DEFAULT_SIZE))
options.append(
'-nw {}'.format(window_size)) # increase size of windows
options.append('--java-mem-size={}'.format(
self.JAVA_MEM_DEFAULT_SIZE))
self.run_cli_command('multi-bamqc', options)
package_info = self.package_output_folder(
workdir, 'QualiMap_report',
'HTML report directory for QualiMap Multi-sample BAM QC',
'multisampleBamQcReport.html')
return {
'qc_result_folder_path': workdir,
'qc_result_zip_info': package_info
}
def get_alignments_from_set(self, alignment_set_ref):
set_data = self.set_api.get_reads_alignment_set_v1({
'ref':
alignment_set_ref,
'include_item_info':
1
})
items = set_data['data']['items']
reads_alignment_data = []
for alignment in items:
alignment_info = self.rau.download_alignment(
{'source_ref': alignment['ref']})
bam_file_path = self.find_my_bam_file(
alignment_info['destination_dir'])
label = None
if 'label' in alignment:
label = alignment['label']
reads_alignment_data.append({
'bam_file_path': bam_file_path,
'ref': alignment['ref'],
'label': label,
'info': alignment['info']
})
return reads_alignment_data
def create_multi_qualimap_cfg(self, reads_alignment_info, workdir):
# Group by labels if there is at least one defined
use_labels = False
for alignment in reads_alignment_info:
if alignment['label']:
use_labels = True
break
# write the file
input_file_path = os.path.join(workdir, 'multi_input.txt')
input_file = open(input_file_path, 'w')
name_lookup = {}
for alignment in reads_alignment_info:
name = alignment['info'][1]
if name in name_lookup:
name_lookup[name] += 1
name = name + '_' + str(name_lookup[name])
else:
name_lookup[name] = 1
input_file.write(name + '\t' + alignment['bam_file_path'])
if use_labels:
if alignment['label']:
input_file.write('\t' + alignment['label'])
else:
input_file.write('\tunlabeled')
input_file.write('\n')
input_file.close()
return input_file_path
def get_run_info(self, params):
info = self.get_obj_info(params['input_ref'])
obj_type = self.get_type_from_obj_info(info)
if obj_type in ['KBaseRNASeq.RNASeqAlignment']:
return {'mode': 'single', 'input_info': info}
if obj_type in [
'KBaseRNASeq.RNASeqAlignmentSet', 'KBaseSets.ReadsAlignmentSet'
]:
return {'mode': 'multi', 'input_info': info}
raise ValueError('Object type of input_ref is not valid, was: ' +
str(obj_type))
def validate_params(self, params):
if 'input_ref' not in params:
raise ValueError(
'required parameter field "input_ref" was not set')
create_report = False
if 'create_report' in params:
if int(params['create_report']) == 1:
if 'output_workspace' not in params:
raise ValueError(
'If "create_report" was set, then "output_workspace" is required'
)
if not params['output_workspace']:
raise ValueError(
'If "create_report" was set, then "output_workspace" is required'
)
create_report = True
params['create_report'] = create_report
def run_cli_command(self, command, options, cwd=None):
if command not in self.valid_commands:
raise ValueError('Invalid QualiMap command: ' + str(command))
command = [self.QUALIMAP_PATH, command] + options
print('Running: ' + ' '.join(command))
if not cwd:
cwd = self.scratch_dir
p = subprocess.Popen(command, cwd=cwd, shell=False)
exitCode = p.wait()
if (exitCode == 0):
print('Success, exit code was: ' + str(exitCode))
else:
raise ValueError('Error running command: ' + ' '.join(command) +
'\n' + 'Exit Code: ' + str(exitCode))
def find_my_bam_file(self, dirpath):
bam_path = None
for f in os.listdir(dirpath):
fullpath = os.path.join(dirpath, f)
if os.path.isfile(fullpath) and f.lower().endswith('.bam'):
if bam_path is not None:
raise ValueError(
'Error! Too many BAM files were downloaded for this alignment!'
)
bam_path = fullpath
if bam_path is None:
raise ValueError(
'Error! No BAM files were downloaded for this alignment!')
return bam_path
def package_output_folder(self, folder_path, zip_file_name,
zip_file_description, index_html_file):
''' Simple utility for packaging a folder and saving to shock '''
output = self.dfu.file_to_shock({
'file_path': folder_path,
'make_handle': 0,
'pack': 'zip'
})
return {
'shock_id': output['shock_id'],
'name': zip_file_name,
'description': zip_file_description,
'index_html_file_name': index_html_file
}
def get_type_from_obj_info(self, info):
return info[2].split('-')[0]
def get_obj_info(self, ref):
return self.ws.get_object_info3({'objects': [{
'ref': ref
}]})['infos'][0]
class RNASeqDownloaderUtils:
def __init__(self, config):
log('--->\nInitializing RNASeqDownloaderUtils instance:\n config: %s' %
config)
self.scratch = config['scratch']
self.callback_url = config['SDK_CALLBACK_URL']
self.token = config['KB_AUTH_TOKEN']
self.dfu = DataFileUtil(self.callback_url, token=self.token)
self.rau = ReadsAlignmentUtils(self.callback_url, token=self.token)
def download_RNASeq(self, params):
"""
download_RNASeq: download RNASeq Alignment/Expression/DifferentialExpression zip file
params:
input_ref: RNASeq object reference ID
rna_seq_type: one of ['RNASeqAlignment',
'RNASeqExpression',
'RNASeqDifferentialExpression']
return:
shock_id: Shock ID of stored zip file
"""
log('--->\nrunning RNASeqDownloaderUtils.download_RNASeq:\nparams: %s'
% params)
# Validate params
self.validate_download_rna_seq_alignment_parameters(params)
# Download RNASeq zip file
# RNASeq Alignemnt, Expression and DifferentialExpression
# has same object_data/handle_data structure
returnVal = self._download_rna_seq_zip(params.get('input_ref'))
return returnVal
def download_RNASeq_Alignment(self, params):
"""
download_RNASeq: download RNASeq Alignment/Expression/DifferentialExpression zip file
params:
input_ref: RNASeq object reference ID
rna_seq_type: 'RNASeqAlignment'
download_file_type: one of 'bam', 'sam' or 'bai'
return:
shock_id: Shock ID of stored zip file
"""
log('--->\nrunning RNASeqDownloaderUtils.download_RNASeq_Alignment:\nparams: %s'
% params)
# Validate params
self.validate_download_rna_seq_alignment_parameters(params)
input_ref = params.get('input_ref')
returnVal = dict()
download_file_type = params.get('download_file_type')
if download_file_type == 'bam':
destination_dir = self.rau.download_alignment({
'source_ref': input_ref,
'downloadBAI': True
})['destination_dir']
shock_id = self._upload_dir_to_shock(destination_dir)
elif download_file_type == 'sam':
destination_dir = self.rau.download_alignment({
'source_ref': input_ref,
'downloadSAM': True,
'downloadBAI': True
})['destination_dir']
files = os.listdir(destination_dir)
bam_files = [x for x in files if re.match('.*\.bam', x)]
for bam_file in bam_files:
log('removing file: {}'.format(bam_file))
os.remove(os.path.join(destination_dir, bam_file))
shock_id = self._upload_dir_to_shock(destination_dir)
returnVal['shock_id'] = shock_id
return returnVal
def validate_download_rna_seq_alignment_parameters(self, params):
"""
validate_download_rna_seq_alignment_parameters:
validates params passed to download_rna_seq_alignment method
"""
# check required parameters
for p in ['input_ref', 'rna_seq_type']:
if p not in params:
raise ValueError('"' + p +
'" parameter is required, but missing')
# check supportive RNASeq types
valid_rnaseq_types = [
'RNASeqAlignment', 'RNASeqExpression',
'RNASeqDifferentialExpression'
]
if params['rna_seq_type'] not in valid_rnaseq_types:
raise ValueError('Unexpected RNASeq type: %s' %
params['rna_seq_type'])
def _download_rna_seq_zip(self, input_ref):
"""
_download_rna_seq_zip: download RNASeq's archive zip file
returns:
shock_id: Shock ID of stored zip file
"""
# get object data
object_data = self._get_object_data(input_ref)
log('---> getting object data\n object_date: %s' %
json.dumps(object_data, indent=1))
# get handle data
handle = self._get_handle_data(object_data)
log('---> getting handle data\n handle data: %s' %
json.dumps(object_data, indent=1))
# make tmp directory for downloading
dstdir = os.path.join(self.scratch, 'tmp')
if not os.path.exists(dstdir):
os.makedirs(dstdir)
# download original zip file and save to tmp directory
handle_id = handle.get('hid')
original_zip_file_path = self._download_original_zip_file(
handle_id, dstdir)
log('---> loading %s to shock' % original_zip_file_path)
shock_id = self._upload_to_shock(original_zip_file_path)
log('---> removing folder: %s' % dstdir)
shutil.rmtree(dstdir)
returnVal = {"shock_id": shock_id}
return returnVal
def _get_object_data(self, input_ref):
"""
_get_object_data: get object_data using DataFileUtil
"""
get_objects_params = {
'object_refs': [input_ref],
'ignore_errors': False
}
object_data = self.dfu.get_objects(get_objects_params)
return object_data
def _get_handle_data(self, object_data):
"""
_get_handle_data: get Handle from object_data
"""
try:
handle = object_data.get('data')[0].get('data').get('file')
except:
error_msg = "Unexpected object format. Refer to DataFileUtil.get_objects definition\n"
error_msg += "object_data:\n%s" % json.dumps(object_data, indent=1)
raise ValueError(error_msg)
if handle is None:
error_msg = "object_data does NOT have Handle(file key)\n"
error_msg += "object_data:\n%s" % json.dumps(object_data, indent=1)
raise ValueError(error_msg)
elif handle.get('hid') is None:
error_msg = "Handle does have NOT HandleId(hid key)\n"
error_msg += "handle_data:\n%s" % json.dumps(handle, indent=1)
raise ValueError(error_msg)
else:
return handle
def _download_original_zip_file(self, handle_id, dstdir):
"""
_download_original_zip_file: download original archive .zip file using DataFileUtil
"""
shock_to_file_params = {'handle_id': handle_id, 'file_path': dstdir}
original_zip_file = self.dfu.shock_to_file(shock_to_file_params)
original_zip_file_path = original_zip_file.get('file_path')
return original_zip_file_path
def _upload_to_shock(self, file_path):
"""
_upload_to_shock: upload target file to shock using DataFileUtil
"""
file_to_shock_params = {'file_path': file_path}
shock_file = self.dfu.file_to_shock(file_to_shock_params)
shock_id = shock_file.get('shock_id')
return shock_id
def _upload_dir_to_shock(self, directory):
"""
_upload_to_shock: upload target file to shock using DataFileUtil
"""
file_to_shock_params = {'file_path': directory, 'pack': 'zip'}
shock_file = self.dfu.file_to_shock(file_to_shock_params)
shock_id = shock_file.get('shock_id')
return shock_id
class QualiMapRunner:
QUALIMAP_PATH = '/kb/module/qualimap-bin/qualimap'
def __init__(self, scratch_dir, callback_url, workspace_url, srv_wiz_url):
self.scratch_dir = scratch_dir
self.rau = ReadsAlignmentUtils(callback_url)
self.kbr = KBaseReport(callback_url)
self.dfu = DataFileUtil(callback_url)
self.set_api = SetAPI(srv_wiz_url)
self.ws = Workspace(workspace_url)
self.valid_commands = ['bamqc', 'multi-bamqc']
def run_app(self, params):
self.validate_params(params)
print('Validated Params = ')
pprint(params)
run_info = self.get_run_info(params)
if run_info['mode'] == 'single':
result = self.run_bamqc(params['input_ref'],
run_info['input_info'])
elif run_info['mode'] == 'multi':
result = self.run_multi_sample_qc(params['input_ref'],
run_info['input_info'])
else:
raise ValueError(
'Error in fetching the type to determine run settings.')
if params['create_report']:
result = self.create_report(result, params['output_workspace'])
return result
def create_report(self, result, output_workspace):
qc_result_zip_info = result['qc_result_zip_info']
report_info = self.kbr.create_extended_report({
'message':
'',
'objects_created': [],
'direct_html_link_index':
0,
'html_links': [{
'shock_id': qc_result_zip_info['shock_id'],
'name': qc_result_zip_info['index_html_file_name'],
'label': qc_result_zip_info['name']
}],
'report_object_name':
'qualimap_report' + str(uuid.uuid4()),
'workspace_name':
output_workspace
})
result['report_name'] = report_info['name']
result['report_ref'] = report_info['ref']
return result
def run_bamqc(self, input_ref, input_info):
# download the input and setup a working dir
alignment_info = self.rau.download_alignment({'source_ref': input_ref})
bam_file_path = self.find_my_bam_file(
alignment_info['destination_dir'])
workdir = os.path.join(self.scratch_dir,
'qualimap_' + str(int(time.time() * 10000)))
options = [
'-bam', bam_file_path, '-outdir', workdir, '-outformat', 'html'
]
self.run_cli_command('bamqc', options)
package_info = self.package_output_folder(
workdir, 'QualiMap_report',
'HTML report directory for QualiMap BAM QC', 'qualimapReport.html')
return {
'qc_result_folder_path': workdir,
'qc_result_zip_info': package_info
}
def run_multi_sample_qc(self, input_ref, input_info):
# download the input and setup a working dir
reads_alignment_info = self.get_alignments_from_set(input_ref)
suffix = 'qualimap_' + str(int(time.time() * 10000))
workdir = os.path.join(self.scratch_dir, suffix)
os.makedirs(workdir)
input_file_path = self.create_multi_qualimap_cfg(
reads_alignment_info, workdir)
options = [
'-d', input_file_path, '-r', '-outdir', workdir, '-outformat',
'html'
]
self.run_cli_command('multi-bamqc', options)
package_info = self.package_output_folder(
workdir, 'QualiMap_report',
'HTML report directory for QualiMap Multi-sample BAM QC',
'multisampleBamQcReport.html')
return {
'qc_result_folder_path': workdir,
'qc_result_zip_info': package_info
}
def get_alignments_from_set(self, alignment_set_ref):
set_data = self.set_api.get_reads_alignment_set_v1({
'ref':
alignment_set_ref,
'include_item_info':
1
})
items = set_data['data']['items']
reads_alignment_data = []
for alignment in items:
alignment_info = self.rau.download_alignment(
{'source_ref': alignment['ref']})
bam_file_path = self.find_my_bam_file(
alignment_info['destination_dir'])
label = None
if 'label' in alignment:
label = alignment['label']
reads_alignment_data.append({
'bam_file_path': bam_file_path,
'ref': alignment['ref'],
'label': label,
'info': alignment['info']
})
return reads_alignment_data
def create_multi_qualimap_cfg(self, reads_alignment_info, workdir):
# Group by labels if there is at least one defined
use_labels = False
for alignment in reads_alignment_info:
if alignment['label']:
use_labels = True
break
# write the file
input_file_path = os.path.join(workdir, 'multi_input.txt')
input_file = open(input_file_path, 'w')
name_lookup = {}
for alignment in reads_alignment_info:
name = alignment['info'][1]
if name in name_lookup:
name_lookup[name] += 1
name = name + '_' + str(name_lookup[name])
else:
name_lookup[name] = 1
input_file.write(name + '\t' + alignment['bam_file_path'])
if use_labels:
if alignment['label']:
input_file.write('\t' + alignment['label'])
else:
input_file.write('\tunlabeled')
input_file.write('\n')
input_file.close()
return input_file_path
def get_run_info(self, params):
info = self.get_obj_info(params['input_ref'])
obj_type = self.get_type_from_obj_info(info)
if obj_type in ['KBaseRNASeq.RNASeqAlignment']:
return {'mode': 'single', 'input_info': info}
if obj_type in [
'KBaseRNASeq.RNASeqAlignmentSet', 'KBaseSets.ReadsAlignmentSet'
]:
return {'mode': 'multi', 'input_info': info}
raise ValueError('Object type of input_ref is not valid, was: ' +
str(obj_type))
def validate_params(self, params):
if 'input_ref' not in params:
raise ValueError(
'required parameter field "input_ref" was not set')
create_report = False
if 'create_report' in params:
if int(params['create_report']) == 1:
if 'output_workspace' not in params:
raise ValueError(
'If "create_report" was set, then "output_workspace" is required'
)
if not params['output_workspace']:
raise ValueError(
'If "create_report" was set, then "output_workspace" is required'
)
create_report = True
params['create_report'] = create_report
def run_cli_command(self, command, options, cwd=None):
if command not in self.valid_commands:
raise ValueError('Invalid QualiMap command: ' + str(command))
command = [self.QUALIMAP_PATH, command] + options
print('Running: ' + ' '.join(command))
if not cwd:
cwd = self.scratch_dir
p = subprocess.Popen(command, cwd=cwd, shell=False)
exitCode = p.wait()
if (exitCode == 0):
print('Success, exit code was: ' + str(exitCode))
else:
raise ValueError('Error running command: ' + ' '.join(command) +
'\n' + 'Exit Code: ' + str(exitCode))
def find_my_bam_file(self, dirpath):
bam_path = None
for f in os.listdir(dirpath):
fullpath = os.path.join(dirpath, f)
if os.path.isfile(fullpath) and f.lower().endswith('.bam'):
if bam_path is not None:
raise ValueError(
'Error! Too many BAM files were downloaded for this alignment!'
)
bam_path = fullpath
if bam_path is None:
raise ValueError(
'Error! No BAM files were downloaded for this alignment!')
return bam_path
def package_output_folder(self, folder_path, zip_file_name,
zip_file_description, index_html_file):
''' Simple utility for packaging a folder and saving to shock '''
output = self.dfu.file_to_shock({
'file_path': folder_path,
'make_handle': 0,
'pack': 'zip'
})
return {
'shock_id': output['shock_id'],
'name': zip_file_name,
'description': zip_file_description,
'index_html_file_name': index_html_file
}
def get_type_from_obj_info(self, info):
return info[2].split('-')[0]
def get_obj_info(self, ref):
return self.ws.get_object_info3({'objects': [{
'ref': ref
}]})['infos'][0]
class CufflinksUtils:
CUFFLINKS_TOOLKIT_PATH = '/opt/cufflinks/'
GFFREAD_TOOLKIT_PATH = '/opt/cufflinks/'
def __init__(self, config):
"""
:param config:
:param logger:
:param directory: Working directory
:param urls: Service urls
"""
# BEGIN_CONSTRUCTOR
self.ws_url = config["workspace-url"]
self.ws_url = config["workspace-url"]
self.callback_url = config['SDK_CALLBACK_URL']
self.srv_wiz_url = config['srv-wiz-url']
self.token = config['KB_AUTH_TOKEN']
self.shock_url = config['shock-url']
self.dfu = DataFileUtil(self.callback_url)
self.gfu = GenomeFileUtil(self.callback_url)
self.au = AssemblyUtil(self.callback_url)
self.rau = ReadsAlignmentUtils(self.callback_url)
self.set_api = SetAPI(self.srv_wiz_url, service_ver='dev')
self.eu = ExpressionUtils(self.callback_url)
self.ws = Workspace(self.ws_url, token=self.token)
self.scratch = os.path.join(config['scratch'], str(uuid.uuid4()))
self._mkdir_p(self.scratch)
self.tool_used = "Cufflinks"
self.tool_version = os.environ['VERSION']
# END_CONSTRUCTOR
pass
def parse_FPKMtracking_calc_TPM(self, filename):
"""
Generates TPM from FPKM
:return:
"""
fpkm_dict = {}
tpm_dict = {}
gene_col = 0
fpkm_col = 9
sum_fpkm = 0.0
with open(filename) as f:
next(f)
for line in f:
larr = line.split("\t")
gene_id = larr[gene_col]
if gene_id != "":
fpkm = float(larr[fpkm_col])
sum_fpkm = sum_fpkm + fpkm
fpkm_dict[gene_id] = math.log(fpkm + 1, 2)
tpm_dict[gene_id] = fpkm
if sum_fpkm == 0.0:
log("Warning: Unable to calculate TPM values as sum of FPKM values is 0"
)
else:
for g in tpm_dict:
tpm_dict[g] = math.log((tpm_dict[g] / sum_fpkm) * 1e6 + 1, 2)
return fpkm_dict, tpm_dict
def _mkdir_p(self, path):
"""
_mkdir_p: make directory for given path
"""
if not path:
return
try:
os.makedirs(path)
except OSError as exc:
if exc.errno == errno.EEXIST and os.path.isdir(path):
pass
else:
raise
def _validate_run_cufflinks_params(self, params):
"""
_validate_run_cufflinks_params:
Raises an exception if params are invalid
"""
log('Start validating run_cufflinks params')
# check for required parameters
for p in ['alignment_object_ref', 'workspace_name', 'genome_ref']:
if p not in params:
raise ValueError(
'"{}" parameter is required, but missing'.format(p))
def _run_command(self, command):
"""
_run_command: run command and print result
"""
log('Start executing command:\n{}'.format(command))
pipe = subprocess.Popen(command, stdout=subprocess.PIPE, shell=True)
output = pipe.communicate()[0]
exitCode = pipe.returncode
if (exitCode == 0):
log('Executed command:\n{}\n'.format(command) +
'Exit Code: {}\nOutput:\n{}'.format(exitCode, output))
else:
error_msg = 'Error running command:\n{}\n'.format(command)
error_msg += 'Exit Code: {}\nOutput:\n{}'.format(exitCode, output)
raise ValueError(error_msg)
def _run_gffread(self, gff_path, gtf_path):
"""
_run_gffread: run gffread script
ref: http://cole-trapnell-lab.github.io/cufflinks/file_formats/#the-gffread-utility
"""
log('converting gff to gtf')
command = self.GFFREAD_TOOLKIT_PATH + '/gffread '
command += "-E {0} -T -o {1}".format(gff_path, gtf_path)
self._run_command(command)
def _create_gtf_annotation_from_genome(self, genome_ref):
"""
Create reference annotation file from genome
"""
ref = self.ws.get_object_subset([{
'ref':
genome_ref,
'included': ['contigset_ref', 'assembly_ref']
}])
if 'contigset_ref' in ref[0]['data']:
contig_id = ref[0]['data']['contigset_ref']
elif 'assembly_ref' in ref[0]['data']:
contig_id = ref[0]['data']['assembly_ref']
if contig_id is None:
raise ValueError(
"Genome at {0} does not have reference to the assembly object".
format(genome_ref))
print(contig_id)
log("Generating GFF file from Genome")
try:
ret = self.au.get_assembly_as_fasta({'ref': contig_id})
output_file = ret['path']
mapping_filename = c_mapping.create_sanitized_contig_ids(
output_file)
os.remove(output_file)
# get the GFF
ret = self.gfu.genome_to_gff({'genome_ref': genome_ref})
genome_gff_file = ret['file_path']
c_mapping.replace_gff_contig_ids(genome_gff_file,
mapping_filename,
to_modified=True)
gtf_ext = ".gtf"
if not genome_gff_file.endswith(gtf_ext):
gtf_path = os.path.splitext(genome_gff_file)[0] + '.gtf'
self._run_gffread(genome_gff_file, gtf_path)
else:
gtf_path = genome_gff_file
log("gtf file : " + gtf_path)
except Exception:
raise ValueError(
"Generating GTF file from Genome Annotation object Failed : {}"
.format("".join(traceback.format_exc())))
return gtf_path
def _get_gtf_file(self, alignment_ref):
"""
_get_gtf_file: get the reference annotation file (in GTF or GFF3 format)
"""
result_directory = self.scratch
alignment_data = self.ws.get_objects2(
{'objects': [{
'ref': alignment_ref
}]})['data'][0]['data']
genome_ref = alignment_data.get('genome_id')
# genome_name = self.ws.get_object_info([{"ref": genome_ref}], includeMetadata=None)[0][1]
# ws_gtf = genome_name+"_GTF_Annotation"
genome_data = self.ws.get_objects2({'objects': [{
'ref': genome_ref
}]})['data'][0]['data']
gff_handle_ref = genome_data.get('gff_handle_ref')
if gff_handle_ref:
log('getting reference annotation file from genome')
annotation_file = self.dfu.shock_to_file({
'handle_id': gff_handle_ref,
'file_path': result_directory,
'unpack': 'unpack'
})['file_path']
else:
annotation_file = self._create_gtf_annotation_from_genome(
genome_ref)
return annotation_file
def _get_gtf_file_from_genome_ref(self, genome_ref):
"""
_get_gtf_file: get the reference annotation file (in GTF or GFF3 format)
"""
result_directory = self.scratch
genome_data = self.ws.get_objects2({'objects': [{
'ref': genome_ref
}]})['data'][0]['data']
gff_handle_ref = genome_data.get('gff_handle_ref')
if gff_handle_ref:
log('getting reference annotation file from genome')
annotation_file = self.dfu.shock_to_file({
'handle_id': gff_handle_ref,
'file_path': result_directory,
'unpack': 'unpack'
})['file_path']
else:
annotation_file = self._create_gtf_annotation_from_genome(
genome_ref)
return annotation_file
def _get_input_file(self, alignment_ref):
"""
_get_input_file: get input BAM file from Alignment object
"""
bam_file_dir = self.rau.download_alignment(
{'source_ref': alignment_ref})['destination_dir']
files = os.listdir(bam_file_dir)
bam_file_list = [
file for file in files if re.match(r'.*\_sorted\.bam', file)
]
if not bam_file_list:
bam_file_list = [
file for file in files if re.match(r'.*(?<!sorted)\.bam', file)
]
if not bam_file_list:
raise ValueError('Cannot find .bam file from alignment {}'.format(
alignment_ref))
bam_file_name = bam_file_list[0]
bam_file = os.path.join(bam_file_dir, bam_file_name)
return bam_file
def _generate_command(self, params):
"""
_generate_command: generate cufflinks command
"""
cufflinks_command = '/opt/cufflinks/cufflinks'
cufflinks_command += (' -q --no-update-check -p ' +
str(params.get('num_threads', 1)))
if 'max_intron_length' in params and params[
'max_intron_length'] is not None:
cufflinks_command += (' --max-intron-length ' +
str(params['max_intron_length']))
if 'min_intron_length' in params and params[
'min_intron_length'] is not None:
cufflinks_command += (' --min-intron-length ' +
str(params['min_intron_length']))
if 'overhang_tolerance' in params and params[
'overhang_tolerance'] is not None:
cufflinks_command += (' --overhang-tolerance ' +
str(params['overhang_tolerance']))
cufflinks_command += " -o {0} -G {1} {2}".format(
params['result_directory'], params['gtf_file'],
params['input_file'])
log('Generated cufflinks command: {}'.format(cufflinks_command))
return cufflinks_command
def _process_rnaseq_alignment_object(self, params):
"""
_process_alignment_object: process KBaseRNASeq.RNASeqAlignment type input object
"""
log('start processing RNASeqAlignment object\nparams:\n{}'.format(
json.dumps(params, indent=1)))
alignment_ref = params.get('alignment_ref')
result_directory = os.path.join(self.scratch, str(uuid.uuid4()))
self._mkdir_p(result_directory)
params['result_directory'] = str(result_directory)
# input files
params['input_file'] = self._get_input_file(alignment_ref)
if not params.get('gtf_file'):
params['gtf_file'] = self._get_gtf_file(alignment_ref)
if '/' not in params['genome_ref']:
params['genome_ref'] = params['workspace_name'] + '/' + params[
'genome_ref']
command = self._generate_command(params)
self._run_command(command)
expression_obj_ref = self._save_rnaseq_expression(
result_directory, alignment_ref, params.get('workspace_name'),
params.get('genome_ref'), params['gtf_file'],
params['expression_suffix'])
returnVal = {
'result_directory': result_directory,
'expression_obj_ref': expression_obj_ref,
'alignment_ref': alignment_ref
}
expression_name = self.ws.get_object_info([{
"ref": expression_obj_ref
}],
includeMetadata=None)[0][1]
widget_params = {
"output": expression_name,
"workspace": params.get('workspace_name')
}
returnVal.update(widget_params)
return returnVal
def _process_kbasesets_alignment_object(self, params):
"""
_process_alignment_object: process KBaseRNASeq.RNASeqAlignment type input object
"""
log('start processing KBaseSets object\nparams:\n{}'.format(
json.dumps(params, indent=1)))
alignment_ref = params.get('alignment_ref')
result_directory = os.path.join(self.scratch, str(uuid.uuid4()))
self._mkdir_p(result_directory)
params['result_directory'] = str(result_directory)
# input files
params['input_file'] = self._get_input_file(alignment_ref)
if not params.get('gtf_file'):
params['gtf_file'] = self._get_gtf_file(alignment_ref)
command = self._generate_command(params)
self._run_command(command)
expression_obj_ref = self._save_kbasesets_expression(
result_directory, alignment_ref, params.get('workspace_name'),
params.get('genome_ref'), params.get('gtf_file'),
params.get('expression_suffix'))
returnVal = {
'result_directory': result_directory,
'expression_obj_ref': expression_obj_ref,
'alignment_ref': alignment_ref
}
expression_name = self.ws.get_object_info([{
"ref": expression_obj_ref
}],
includeMetadata=None)[0][1]
widget_params = {
"output": expression_name,
"workspace": params.get('workspace_name')
}
returnVal.update(widget_params)
return returnVal
def _generate_html_report(self, result_directory, obj_ref):
"""
_generate_html_report: generate html summary report
"""
log('Start generating html report')
html_report = list()
output_directory = os.path.join(self.scratch, str(uuid.uuid4()))
self._mkdir_p(output_directory)
result_file_path = os.path.join(output_directory, 'report.html')
expression_object = self.ws.get_objects2(
{'objects': [{
'ref': obj_ref
}]})['data'][0]
expression_object_type = expression_object.get('info')[2]
Overview_Content = ''
if re.match('KBaseRNASeq.RNASeqExpression-\d.\d',
expression_object_type):
Overview_Content += '<p>Generated Expression Object:</p><p>{}</p>'.format(
expression_object.get('info')[1])
elif re.match('KBaseRNASeq.RNASeqExpressionSet-\d.\d',
expression_object_type):
Overview_Content += '<p>Generated Expression Set Object:</p><p>{}</p>'.format(
expression_object.get('info')[1])
Overview_Content += '<br><p>Generated Expression Object:</p>'
for expression_ref in expression_object['data'][
'sample_expression_ids']:
expression_name = self.ws.get_object_info(
[{
"ref": expression_ref
}], includeMetadata=None)[0][1]
Overview_Content += '<p>{}</p>'.format(expression_name)
elif re.match('KBaseSets.ExpressionSet-\d.\d', expression_object_type):
pprint(expression_object)
Overview_Content += '<p>Generated Expression Set Object:</p><p>{}</p>'.format(
expression_object.get('info')[1])
Overview_Content += '<br><p>Generated Expression Object:</p>'
for expression_ref in expression_object['data']['items']:
expression_name = self.ws.get_object_info(
[{
"ref": expression_ref['ref']
}], includeMetadata=None)[0][1]
condition = expression_ref['label']
Overview_Content += '<p>condition:{0}; expression_name: {1}</p>'.format(
condition, expression_name)
with open(result_file_path, 'w') as result_file:
with open(
os.path.join(os.path.dirname(__file__),
'report_template.html'),
'r') as report_template_file:
report_template = report_template_file.read()
report_template = report_template.replace(
'<p>Overview_Content</p>', Overview_Content)
result_file.write(report_template)
html_report.append({
'path':
result_file_path,
'name':
os.path.basename(result_file_path),
'label':
os.path.basename(result_file_path),
'description':
'HTML summary report for Cufflinks App'
})
return html_report
def _save_rnaseq_expression(self, result_directory, alignment_ref,
workspace_name, genome_ref, gtf_file,
expression_suffix):
"""
_save_rnaseq_expression: save Expression object to workspace
"""
log('start saving Expression object')
alignment_object_name = self.ws.get_object_info(
[{
"ref": alignment_ref
}], includeMetadata=None)[0][1]
# set expression name
if re.match('.*_[Aa]lignment$', alignment_object_name):
expression_name = re.sub('_[Aa]lignment$', expression_suffix,
alignment_object_name)
else: # assume user specified suffix
expression_name = alignment_object_name + expression_suffix
expression_ref = self.eu.upload_expression({
'destination_ref':
workspace_name + '/' + expression_name,
'source_dir':
result_directory,
'alignment_ref':
alignment_ref,
'tool_used':
self.tool_used,
'tool_version':
self.tool_version
})['obj_ref']
return expression_ref
def _save_kbasesets_expression(self, result_directory, alignment_ref,
workspace_name, genome_ref, gtf_file,
expression_suffix):
"""
_save_kbasesets_expression: save Expression object to workspace using ExpressionUtils
and SetAPI
"""
log('start saving Expression object')
alignment_info = self.ws.get_object_info3(
{'objects': [{
"ref": alignment_ref
}]})
alignment_object_name = alignment_info['infos'][0][1]
# set expression name
if re.match('.*_[Aa]lignment$', alignment_object_name):
expression_name = re.sub('_[Aa]lignment$', expression_suffix,
alignment_object_name)
else: # assume user specified suffix
expression_name = alignment_object_name + expression_suffix
expression_ref = self.eu.upload_expression({
'destination_ref':
workspace_name + '/' + expression_name,
'source_dir':
result_directory,
'alignment_ref':
alignment_ref,
'tool_used':
self.tool_used,
'tool_version':
self.tool_version
})['obj_ref']
return expression_ref
def _save_rnaseq_expression_set(self, alignment_expression_map,
alignment_set_ref, workspace_name,
expression_set_name):
"""
_save_rnaseq_expression_set: save ExpressionSet object to workspace
"""
log('start saving ExpressionSet object')
if isinstance(workspace_name, int) or workspace_name.isdigit():
workspace_id = workspace_name
else:
workspace_id = self.dfu.ws_name_to_id(workspace_name)
expression_set_data = self._generate_expression_set_data(
alignment_expression_map, alignment_set_ref, expression_set_name)
object_type = 'KBaseRNASeq.RNASeqExpressionSet'
save_object_params = {
'id':
workspace_id,
'objects': [{
'type': object_type,
'data': expression_set_data,
'name': expression_set_name
}]
}
dfu_oi = self.dfu.save_objects(save_object_params)[0]
expression_set_ref = str(dfu_oi[6]) + '/' + str(dfu_oi[0]) + '/' + str(
dfu_oi[4])
return expression_set_ref
def _save_kbasesets_expression_set(self, alignment_expression_map,
alignment_set_ref, workspace_name,
expression_set_name):
"""
_save_kbasesets_expression_set: save ExpressionSet object to workspace
"""
log('start saving ExpressionSet object')
if isinstance(workspace_name, int) or workspace_name.isdigit():
workspace_id = workspace_name
else:
workspace_id = self.dfu.ws_name_to_id(workspace_name)
expression_set_data = self._generate_expression_set_data(
alignment_expression_map, alignment_set_ref, expression_set_name)
object_type = 'KBaseRNASeq.RNASeqExpressionSet'
save_object_params = {
'id':
workspace_id,
'objects': [{
'type': object_type,
'data': expression_set_data,
'name': expression_set_name
}]
}
dfu_oi = self.dfu.save_objects(save_object_params)[0]
expression_set_ref = str(dfu_oi[6]) + '/' + str(dfu_oi[0]) + '/' + str(
dfu_oi[4])
return expression_set_ref
def _generate_report(self,
obj_ref,
workspace_name,
result_directory,
exprMatrix_FPKM_ref=None,
exprMatrix_TPM_ref=None):
"""
_generate_report: generate summary report
"""
log('creating report')
output_files = self._generate_output_file_list(result_directory)
output_html_files = self._generate_html_report(result_directory,
obj_ref)
expression_object = self.ws.get_objects2(
{'objects': [{
'ref': obj_ref
}]})['data'][0]
expression_info = expression_object['info']
expression_data = expression_object['data']
expression_object_type = expression_info[2]
if re.match('KBaseRNASeq.RNASeqExpression-\d+.\d+',
expression_object_type):
objects_created = [{
'ref':
obj_ref,
'description':
'Expression generated by Cufflinks'
}]
elif re.match('KBaseRNASeq.RNASeqExpressionSet-\d+.\d+',
expression_object_type):
objects_created = [{
'ref':
obj_ref,
'description':
'Expression generated by Cufflinks'
}]
elif re.match('KBaseSets.ExpressionSet-\d+.\d+',
expression_object_type):
objects_created = [{
'ref':
obj_ref,
'description':
'ExpressionSet generated by Cufflinks'
}]
items = expression_data['items']
for item in items:
objects_created.append({
'ref':
item['ref'],
'description':
'Expression generated by Cufflinks'
})
objects_created.append({
'ref':
exprMatrix_FPKM_ref,
'description':
'FPKM ExpressionMatrix generated by Cufflinks'
})
objects_created.append({
'ref':
exprMatrix_TPM_ref,
'description':
'TPM ExpressionMatrix generated by Cufflinks'
})
report_params = {
'message': '',
'workspace_name': workspace_name,
'file_links': output_files,
'objects_created': objects_created,
'html_links': output_html_files,
'direct_html_link_index': 0,
'html_window_height': 366,
'report_object_name': 'kb_cufflinks_report_' + str(uuid.uuid4())
}
kbase_report_client = KBaseReport(self.callback_url, token=self.token)
output = kbase_report_client.create_extended_report(report_params)
report_output = {
'report_name': output['name'],
'report_ref': output['ref']
}
return report_output
def _parse_FPKMtracking(self, filename, metric):
result = {}
pos1 = 0
if metric == 'FPKM':
pos2 = 7
if metric == 'TPM':
pos2 = 8
with open(filename) as f:
next(f)
for line in f:
larr = line.split("\t")
if larr[pos1] != "":
try:
result[larr[pos1]] = math.log(float(larr[pos2]) + 1, 2)
except ValueError:
result[larr[pos1]] = math.log(1, 2)
return result
def _generate_output_file_list(self, result_directory):
"""
_generate_output_file_list: zip result files and generate file_links for report
"""
log('Start packing result files')
output_files = list()
output_directory = os.path.join(self.scratch, str(uuid.uuid4()))
self._mkdir_p(output_directory)
result_file = os.path.join(output_directory, 'cufflinks_result.zip')
with zipfile.ZipFile(result_file,
'w',
zipfile.ZIP_DEFLATED,
allowZip64=True) as zip_file:
for root, dirs, files in os.walk(result_directory):
for file in files:
if not (file.endswith('.DS_Store')):
zip_file.write(
os.path.join(root, file),
os.path.join(os.path.basename(root), file))
output_files.append({
'path': result_file,
'name': os.path.basename(result_file),
'label': os.path.basename(result_file),
'description': 'File(s) generated by Cufflinks App'
})
return output_files
def _generate_expression_data(self, result_directory, alignment_ref,
gtf_file, workspace_name, expression_suffix):
"""
_generate_expression_data: generate Expression object with cufflinks output files
"""
alignment_data_object = self.ws.get_objects2(
{'objects': [{
'ref': alignment_ref
}]})['data'][0]
# set expression name
alignment_object_name = alignment_data_object['info'][1]
if re.match('.*_[Aa]lignment$', alignment_object_name):
expression_name = re.sub('_[Aa]lignment$', expression_suffix,
alignment_object_name)
else: # assume user specified suffix
expression_name = alignment_object_name + expression_suffix
expression_data = {
'id': expression_name,
'type': 'RNA-Seq',
'numerical_interpretation': 'FPKM',
'processing_comments': 'log2 Normalized',
'tool_used': self.tool_used,
'tool_version': self.tool_version
}
alignment_data = alignment_data_object['data']
condition = alignment_data.get('condition')
expression_data.update({'condition': condition})
genome_id = alignment_data.get('genome_id')
expression_data.update({'genome_id': genome_id})
read_sample_id = alignment_data.get('read_sample_id')
expression_data.update(
{'mapped_rnaseq_alignment': {
read_sample_id: alignment_ref
}})
exp_dict, tpm_exp_dict = self.parse_FPKMtracking_calc_TPM(
os.path.join(result_directory, 'genes.fpkm_tracking'))
expression_data.update({'expression_levels': exp_dict})
expression_data.update({'tpm_expression_levels': tpm_exp_dict})
handle = self.dfu.file_to_shock({
'file_path': result_directory,
'pack': 'zip',
'make_handle': True
})['handle']
expression_data.update({'file': handle})
return expression_data
def _generate_expression_set_data(self, alignment_expression_map,
alignment_set_ref, expression_set_name):
"""
_generate_expression_set_data: generate ExpressionSet object with cufflinks output files
"""
alignment_set_data_object = self.ws.get_objects2(
{'objects': [{
'ref': alignment_set_ref
}]})['data'][0]
alignment_set_data = alignment_set_data_object['data']
expression_set_data = {
'tool_used': self.tool_used,
'tool_version': self.tool_version,
'id': expression_set_name,
'alignmentSet_id': alignment_set_ref,
'genome_id': alignment_set_data.get('genome_id'),
'sampleset_id': alignment_set_data.get('sampleset_id')
}
sample_expression_ids = []
mapped_expression_objects = []
mapped_expression_ids = []
for alignment_expression in alignment_expression_map:
alignment_ref = alignment_expression.get('alignment_ref')
expression_ref = alignment_expression.get('expression_obj_ref')
sample_expression_ids.append(expression_ref)
mapped_expression_ids.append({alignment_ref: expression_ref})
alignment_name = self.ws.get_object_info(
[{
"ref": alignment_ref
}], includeMetadata=None)[0][1]
expression_name = self.ws.get_object_info(
[{
"ref": expression_ref
}], includeMetadata=None)[0][1]
mapped_expression_objects.append({alignment_name: expression_name})
expression_set_data['sample_expression_ids'] = sample_expression_ids
expression_set_data[
'mapped_expression_objects'] = mapped_expression_objects
expression_set_data['mapped_expression_ids'] = mapped_expression_ids
return expression_set_data
def _process_alignment_set_object(self, params, alignment_object_type):
"""
_process_alignment_set_object: process KBaseRNASeq.RNASeqAlignmentSet type input object
and KBaseSets.ReadsAlignmentSet type object
"""
log('start processing KBaseRNASeq.RNASeqAlignmentSet object or KBaseSets.ReadsAlignmentSet object'
'\nparams:\n{}'.format(json.dumps(params, indent=1)))
alignment_set_ref = params.get('alignment_set_ref')
if re.match('^KBaseRNASeq.RNASeqAlignmentSet-\d*',
alignment_object_type):
params['gtf_file'] = self._get_gtf_file(alignment_set_ref)
else:
if not '/' in params['genome_ref']:
params['genome_ref'] = params['workspace_name'] + '/' + params[
'genome_ref']
params['gtf_file'] = self._get_gtf_file_from_genome_ref(
params['genome_ref'])
alignment_set = self.set_api.get_reads_alignment_set_v1({
'ref':
alignment_set_ref,
'include_item_info':
0,
'include_set_item_ref_paths':
1
})
mul_processor_params = []
for alignment in alignment_set["data"]["items"]:
alignment_ref = alignment['ref_path']
alignment_upload_params = params.copy()
alignment_upload_params['alignment_ref'] = alignment_ref
mul_processor_params.append(alignment_upload_params)
# use the following when you want to run the cmd sequentially
# self._process_kbasesets_alignment_object(mul_processor_params[0])
cpus = min(params.get('num_threads'), multiprocessing.cpu_count())
pool = Pool(ncpus=cpus)
log('running _process_alignment_object with {} cpus'.format(cpus))
alignment_expression_map = pool.map(
self._process_kbasesets_alignment_object, mul_processor_params)
result_directory = os.path.join(self.scratch, str(uuid.uuid4()))
self._mkdir_p(result_directory)
expression_items = list()
for proc_alignment_return in alignment_expression_map:
expression_obj_ref = proc_alignment_return.get(
'expression_obj_ref')
alignment_ref = proc_alignment_return.get('alignment_ref')
alignment_info = self.ws.get_object_info3({
'objects': [{
"ref": alignment_ref
}],
'includeMetadata':
1
})
condition = alignment_info['infos'][0][10]['condition']
expression_items.append({
"ref": expression_obj_ref,
"label": condition,
})
expression_name = self.ws.get_object_info(
[{
"ref": expression_obj_ref
}], includeMetadata=None)[0][1]
self._run_command('cp -R {} {}'.format(
proc_alignment_return.get('result_directory'),
os.path.join(result_directory, expression_name)))
expression_set = {
"description": "generated by kb_cufflinks",
"items": expression_items
}
expression_set_info = self.set_api.save_expression_set_v1({
"workspace":
params['workspace_name'],
"output_object_name":
params['expression_set_name'],
"data":
expression_set
})
returnVal = {
'result_directory': result_directory,
'expression_obj_ref': expression_set_info['set_ref']
}
widget_params = {
"output": params.get('expression_set_name'),
"workspace": params.get('workspace_name')
}
returnVal.update(widget_params)
return returnVal
def _generate_output_object_name(self, params, alignment_object_type,
alignment_object_name):
"""
Generates the output object name based on input object type and name and stores it in
params with key equal to 'expression' or 'expression_set' based on whether the input
object is an alignment or alignment_set.
:param params: module input params
:param alignment_object_type: input alignment object type
:param alignment_object_name: input alignment object name
:param alignment_object_data: input alignment object data
"""
expression_set_suffix = params['expression_set_suffix']
expression_suffix = params['expression_suffix']
if re.match('^KBaseRNASeq.RNASeqAlignment-\d*', alignment_object_type):
if re.match('.*_[Aa]lignment$', alignment_object_name):
params['expression_name'] = re.sub('_[Aa]lignment$',
expression_suffix,
alignment_object_name)
else: # assume user specified suffix
params[
'expression_name'] = alignment_object_name + expression_suffix
if re.match('^KBaseRNASeq.RNASeqAlignmentSet-\d*',
alignment_object_type):
if re.match('.*_[Aa]lignment_[Ss]et$', alignment_object_name):
# set expression set name
params['expression_set_name'] = re.sub('_[Aa]lignment_[Ss]et$',
expression_set_suffix,
alignment_object_name)
else: # assume user specified suffix
params[
'expression_set_name'] = alignment_object_name + expression_set_suffix
if re.match('^KBaseSets.ReadsAlignmentSet-\d*', alignment_object_type):
if re.match('.*_[Aa]lignment_[Ss]et$', alignment_object_name):
# set expression set name
params['expression_set_name'] = re.sub('_[Aa]lignment_[Ss]et$',
expression_set_suffix,
alignment_object_name)
else: # assume user specified suffix
params[
'expression_set_name'] = alignment_object_name + expression_set_suffix
def _save_expression_matrix(self, expressionset_ref, workspace_name):
"""
_save_expression_matrix: save FPKM and TPM ExpressionMatrix
"""
log('start saving ExpressionMatrix object')
expression_set_name = self.ws.get_object_info(
[{
"ref": expressionset_ref
}], includeMetadata=None)[0][1]
output_obj_name_prefix = re.sub('_*[Ee]xpression_*[Ss]et', '',
expression_set_name)
upload_expression_matrix_params = {
'expressionset_ref': expressionset_ref,
'output_obj_name': output_obj_name_prefix,
'workspace_name': workspace_name
}
expression_matrix_refs = self.eu.get_expressionMatrix(
upload_expression_matrix_params)
return expression_matrix_refs
def run_cufflinks_app(self, params):
log('--->\nrunning CufflinksUtil.run_cufflinks_app\n' +
'params:\n{}'.format(json.dumps(params, indent=1)))
self._validate_run_cufflinks_params(params)
alignment_object_ref = params.get('alignment_object_ref')
alignment_object_info = self.ws.get_object_info3(
{"objects": [{
"ref": alignment_object_ref
}]})['infos'][0]
alignment_object_type = alignment_object_info[2]
alignment_object_name = alignment_object_info[1]
# get output object name
self._generate_output_object_name(params, alignment_object_type,
alignment_object_name)
log('--->\nalignment object type: \n' +
'{}'.format(alignment_object_type))
if re.match('^KBaseRNASeq.RNASeqAlignment-\d*', alignment_object_type):
params.update({'alignment_ref': alignment_object_ref})
returnVal = self._process_rnaseq_alignment_object(params)
report_output = self._generate_report(
returnVal.get('expression_obj_ref'),
params.get('workspace_name'),
returnVal.get('result_directory'))
returnVal.update(report_output)
elif re.match('^KBaseRNASeq.RNASeqAlignmentSet-\d*', alignment_object_type) or \
re.match('^KBaseSets.ReadsAlignmentSet-\d*', alignment_object_type):
params.update({'alignment_set_ref': alignment_object_ref})
returnVal = self._process_alignment_set_object(
params, alignment_object_type)
expression_matrix_refs = self._save_expression_matrix(
returnVal['expression_obj_ref'], params.get('workspace_name'))
returnVal.update(expression_matrix_refs)
report_output = self._generate_report(
returnVal['expression_obj_ref'], params.get('workspace_name'),
returnVal['result_directory'],
expression_matrix_refs['exprMatrix_FPKM_ref'],
expression_matrix_refs['exprMatrix_TPM_ref'])
returnVal.update(report_output)
else:
raise ValueError(
'None RNASeqAlignment type\nObject info:\n{}'.format(
alignment_object_info))
return returnVal