def delete(self, id):
"""Delete pipeline from NGSOnto
This method allows removing a pipeline from a project.
Parameters
----------
id: str
project identifier
Returns
-------
"""
args = pipeline_delete_parser.parse_args()
pipeline_id = args.pipeline_id
studyURI = dbconAg.createURI(namespace=localNSpace + "projects/",
localname=str(id))
hasPart = dbconAg.createURI(namespace=obo, localname="OBI_0000471")
pipelineURI = dbconAg.createURI(namespace=localNSpace + "projects/",
localname=str(id) + "/pipelines/" +
str(pipeline_id))
dbconAg.remove(studyURI, hasPart, pipelineURI)
return 204
def put(self, id, id2, id3):
"""Change a specific output
(DEPRECATED)
Parameters
----------
id
id2
id3
Returns
-------
"""
args = process_put_output_parser.parse_args()
output_prop_to_type = {
"run_info": "NGS_0000092",
"run_output": "NGS_0000093",
"run_stats": "NGS_0000094",
"log_file": "NGS_0000096",
"status": "NGS_0000097"
}
try:
processURI = dbconAg.createURI(namespace=localNSpace + "projects/",
localname=str(id) + "/pipelines/" +
str(id2) + "/processes/" + str(id3))
# get output URI from process
hasOutput = dbconAg.createURI(namespace=obo,
localname="RO_0002234")
statements = dbconAg.getStatements(processURI, hasOutput, None)
outputURI = parseAgraphStatementsRes(statements)
statements.close()
outputURI = dbconAg.createURI(outputURI[0]['obj'])
runInfo = dbconAg.createLiteral((args.property),
datatype=XMLSchema.STRING)
runInfoProp = dbconAg.createURI(
namespace=obo, localname=output_prop_to_type[args.property])
dbconAg.remove(outputURI, runInfoProp, None)
# add outputs paths to process
stmt1 = dbconAg.createStatement(outputURI, runInfoProp, runInfo)
# send to allegro
dbconAg.add(stmt1)
return 202
except Exception as e:
print e
return 404
def post(self, id, id2):
"""Add job identifier to process
This method allows adding a job identifier to a process. It requires
the processes ids and the job identifiers.
Parameters
----------
id: str
project identifier
id2: str
pipeline identifier
Returns
-------
dict: status and job identifiers
"""
args = parser_jobid.parse_args()
tasks = args.task_ids.split(',')
processes = args.processes_ids.split(',')
countadded = 0
for index in range(0, len(processes)):
try:
processURI = dbconAg.createURI(
namespace=localNSpace + "projects/",
localname=str(id) + "/pipelines/" + str(id2) +
"/processes/" + str(processes[index]))
indexProp = dbconAg.createURI(namespace=obo,
localname="NGS_0000089")
indexInt = dbconAg.createLiteral(tasks[index],
datatype=XMLSchema.STRING)
# add jobID to process
dbconAg.remove(processURI, indexProp, None)
stmt1 = dbconAg.createStatement(processURI, indexProp,
indexInt)
# send to allegro
dbconAg.add(stmt1)
countadded += 1
except Exception:
print 'error mapping process'
if countadded == len(tasks):
return {'status': 202, 'tasks': tasks}
else:
return 404
def set_unique_prop_output(project_id, pipeline_id, process_id,
property_type, property_value):
output_prop_to_type = {
"run_info": "NGS_0000092", "run_output": "NGS_0000093",
"warnings": "NGS_0000094", "log_file": "NGS_0000096",
"status": "NGS_0000097"
}
property_types = property_type.split(",")
property_values = property_value.split(",")
try:
for p, v in zip(property_types, property_values):
# Agraph
processURI = dbconAg.createURI(
namespace=localNSpace + "projects/",
localname=str(project_id) + "/pipelines/" + str(
pipeline_id) + "/processes/" + str(process_id))
# get output URI from process
hasOutput = dbconAg.createURI(namespace=obo,
localname="RO_0002234")
statements = dbconAg.getStatements(processURI,
hasOutput, None)
outputURI = parseAgraphStatementsRes(statements)
statements.close()
outputURI = dbconAg.createURI(outputURI[0]['obj'])
runInfo = dbconAg.createLiteral((v),
datatype=XMLSchema.STRING)
runInfoProp = dbconAg.createURI(
namespace=obo, localname=output_prop_to_type[p])
if p != "log_file" and p != "warnings":
dbconAg.remove(outputURI, runInfoProp, None)
# add outputs paths to process
stmt1 = dbconAg.createStatement(outputURI, runInfoProp,
runInfo)
# send to allegro
dbconAg.add(stmt1)
except Exception as e:
print e
def post(self, id, id2, id3):
"""Add output to process
This method allows adding logs, run information, status and
statistics to the NGSOnto entry.
Parameters
----------
id: str
project identifier
id2: str
pipeline identifier
id3: str
process identifier
Returns
-------
code: status code of the request
"""
args = process_post_output_parser.parse_args()
try:
processURI = dbconAg.createURI(namespace=localNSpace + "projects/",
localname=str(id) + "/pipelines/" +
str(id2) + "/processes/" + str(id3))
# get output URI from process
hasOutput = dbconAg.createURI(namespace=obo,
localname="RO_0002234")
statements = dbconAg.getStatements(processURI, hasOutput, None)
outputURI = parseAgraphStatementsRes(statements)
statements.close()
outputURI = dbconAg.createURI(outputURI[0]['obj'])
runInfo = dbconAg.createLiteral((args.run_info),
datatype=XMLSchema.STRING)
runInfoProp = dbconAg.createURI(namespace=obo,
localname="NGS_0000092")
runStats = dbconAg.createLiteral((args.output),
datatype=XMLSchema.STRING)
runStatsProp = dbconAg.createURI(namespace=obo,
localname="NGS_0000093")
runFile = dbconAg.createLiteral((args.run_stats),
datatype=XMLSchema.STRING)
runFileProp = dbconAg.createURI(namespace=obo,
localname="NGS_0000094")
runStatus = dbconAg.createLiteral((args.status),
datatype=XMLSchema.STRING)
runStatusProp = dbconAg.createURI(namespace=obo,
localname="NGS_0000097")
dbconAg.remove(outputURI, runInfoProp, None)
dbconAg.remove(outputURI, runStatsProp, None)
dbconAg.remove(outputURI, runFileProp, None)
dbconAg.remove(outputURI, runStatusProp, None)
# add outputs paths to process
stmt1 = dbconAg.createStatement(outputURI, runInfoProp, runInfo)
stmt2 = dbconAg.createStatement(outputURI, runStatsProp, runStats)
stmt3 = dbconAg.createStatement(outputURI, runFileProp, runFile)
stmt4 = dbconAg.createStatement(processURI, runStatusProp,
runStatus)
# send to allegro
dbconAg.add(stmt1)
dbconAg.add(stmt2)
dbconAg.add(stmt3)
dbconAg.add(stmt4)
return 202
except Exception as e:
print e
return 404
def post(self, id, id2):
"""Add processes to pipeline
This method allows adding processes to a pipeline by linking the
protocol ids with the processes.
It requires the project id, pipeline id, protocol ids and strain
identifier.
Parameters
----------
id: str
project identifier
id2: str
pipeline identifier
Returns
-------
list: list of processes identifiers
"""
args = process_post_parser.parse_args()
pipelineStr = localNSpace + "projects/" + str(
id) + "/pipelines/" + str(id2)
# get number of processes already mapped on the pipeline
hasPart = dbconAg.createURI(namespace=obo, localname="BFO_0000051")
pipelineURI = dbconAg.createURI(pipelineStr)
statements = dbconAg.getStatements(pipelineURI, hasPart, None)
jsonResult = parseAgraphStatementsRes(statements)
statements.close()
numberOfProcesses = len(jsonResult)
print "Request 1", str(id2)
# get all ordered workflows from pipeline
queryString = "SELECT (str(?proc) " \
"as ?StrProc) (str(?index) as ?StrIndex)" \
" WHERE{<"+pipelineStr+"> obo:BFO_0000051 ?proc." \
" ?proc obo:NGS_0000081 ?index.}"
tupleQuery = dbconAg.prepareTupleQuery(QueryLanguage.SPARQL,
queryString)
result = tupleQuery.evaluate()
procJsonResult = parseAgraphQueryRes(result, ["StrProc", "StrIndex"])
result.close()
numberOfProcesses = len(procJsonResult)
print "Request 2", str(id2)
# get all ordered workflows from pipeline
queryString = "SELECT ?execStep ?stepIndex" \
" ?workflowURI ?execStep " \
"WHERE {<"+pipelineStr+"> obo:NGS_0000076 ?execStep." \
" ?execStep obo:NGS_0000079" \
" ?workflowURI; obo:NGS_0000081" \
" ?stepIndex3} ORDER BY" \
" ASC(?stepIndex)"
tupleQuery = dbconAg.prepareTupleQuery(QueryLanguage.SPARQL,
queryString)
result = tupleQuery.evaluate()
jsonResult = parseAgraphQueryRes(
result, ["stepIndex", "workflowURI", "execStep"])
result.close()
print "Request 3", str(id2)
# get all protocols per workflow
listOrderedProtocolsURI = []
listOrderedProcessTypes = []
listOrderedMessageTypes = []
for result in jsonResult:
workflowURI = result["workflowURI"]
queryString = "SELECT ?protocStep ?stepIndex" \
" ?protocolURI ?type " \
"WHERE {"+workflowURI+\
" obo:NGS_0000078 ?protocStep. ?protocStep" \
" obo:NGS_0000077 ?protocolURI; obo:NGS_0000081" \
" ?stepIndex. ?protocolURI a ?type. ?type rdfs:label"\
" ?typelabel.} ORDER BY ASC(?stepIndex)"
tupleQuery = dbconAg.prepareTupleQuery(QueryLanguage.SPARQL,
queryString)
result3 = tupleQuery.evaluate()
jsonResult2 = parseAgraphQueryRes(
result3, ["stepIndex", "protocolURI", "type"])
result3.close()
for results in jsonResult2:
for k, v in protocolsTypes.items():
if v in results["type"]:
listOrderedProtocolsURI.append(results["protocolURI"])
listOrderedProcessTypes.append(processTypes[k])
listOrderedMessageTypes.append(processMessages[k])
print "Request 4 all protocols", str(id2)
# Starts at 500 in case does not exists
messageid = 500
# TEST query string
queryString = """SELECT ?index {?message rdf:type/rdfs:subClassOf*
obo:NGS_0000061; obo:NGS_0000081 ?index} order by
desc(?index) limit 1"""
tupleQuery = dbconAg.prepareTupleQuery(QueryLanguage.SPARQL,
queryString)
result = tupleQuery.evaluate()
for bindingSet in result:
messageid = int(str(bindingSet[0]).split('"')[1])
print "Request 5", str(id2)
result.close()
if args.strain_id != "null":
strainid = args.strain_id
rpipid = args.real_pipeline_id
ppipid = rpipid
ppropid = id
pprocid = 0
else:
ppipid = args.parent_pipeline_id
ppropid = args.parent_project_id
pprocid = args.parent_process_id
rpipid = args.real_pipeline_id
if ppipid == rpipid:
for proc_json in procJsonResult:
if int(proc_json["StrIndex"].replace('"', '')) > int(pprocid):
todelUri = dbconAg.createURI(
"<" + proc_json["StrProc"].replace('"', "") + ">")
hasOutputRel = dbconAg.createURI(namespace=obo,
localname="RO_0002234")
statements = dbconAg.getStatements(todelUri, hasOutputRel,
None)
jsonResult = parseAgraphStatementsRes(statements)
statements.close()
todelUri2 = jsonResult[0]["obj"]
todelUri2 = dbconAg.createURI(todelUri2)
dbconAg.remove(todelUri2, None, None)
dbconAg.remove(todelUri, None, None)
dbconAg.remove(None, None, todelUri)
statements = dbconAg.getStatements(todelUri, None, None)
jsonResult = parseAgraphStatementsRes(statements)
statements.close()
numberOfProcesses -= 1
print "Request 6", str(id2)
try:
addedProcesses = numberOfProcesses
hasOutputRel = dbconAg.createURI(namespace=obo,
localname="RO_0002234")
hasInputRel = dbconAg.createURI(namespace=obo,
localname="RO_0002233")
index = dbconAg.createURI(namespace=obo, localname="NGS_0000081")
isRunOfProtocl = dbconAg.createURI(namespace=obo,
localname="NGS_0000091")
# prev process to link (strain URI most of times)
if args.strain_id != "null":
prevMessageURI = dbconAg.createURI(
namespace=localNSpace,
localname="strains/strain_" + str(strainid))
strainTypeURI = dbconAg.createURI(
'http://rdf.ebi.ac.uk/terms/biosd/Sample')
dbconAg.add(prevMessageURI, RDF.TYPE, strainTypeURI)
processes_ids = []
processid = addedProcesses
# Case new run
while addedProcesses < len(listOrderedProcessTypes):
processid += 1
messageid += 1
processURI = dbconAg.createURI(
namespace=localNSpace + "projects/",
localname=str(id) + "/pipelines/" + str(id2) +
"/processes/" + str(processid))
messageURI = dbconAg.createURI(
namespace=localNSpace + "projects/",
localname=str(id) + "/pipelines/" + str(id2) +
"/messages/" + str(messageid))
processTypeURI = dbconAg.createURI(
listOrderedProcessTypes[addedProcesses])
messageTypeURI = dbconAg.createURI(
listOrderedMessageTypes[addedProcesses])
protocolTypeURI = dbconAg.createURI(
listOrderedProtocolsURI[addedProcesses])
indexProp = dbconAg.createURI(namespace=obo,
localname="NGS_0000081")
indexInt = dbconAg.createLiteral((addedProcesses + 1),
datatype=XMLSchema.INT)
messageindexInt = dbconAg.createLiteral((messageid),
datatype=XMLSchema.INT)
# get specific process input type and uri
queryString = """SELECT (STR(?out) as ?messageURI) WHERE
{<"""+localNSpace+"projects/"+str(id)+"/pipelines/"+\
str(rpipid)+"""> obo:BFO_0000051 ?proc. ?proc
obo:NGS_0000081 ?index; obo:RO_0002234 ?out}
order by desc(?out)"""
print queryString
tupleQuery = dbconAg.prepareTupleQuery(QueryLanguage.SPARQL,
queryString)
result5 = tupleQuery.evaluate()
jsonResult2 = parseAgraphQueryRes(result5, ["messageURI"])
result5.close()
for results in jsonResult2:
prevMessageURI = dbconAg.createURI(
results["messageURI"].replace('"', ''))
break
# add process and link to pipeline
dbconAg.add(processURI, RDF.TYPE, processTypeURI)
dbconAg.add(pipelineURI, hasPart, processURI)
stmt1 = dbconAg.createStatement(processURI, indexProp,
indexInt)
dbconAg.add(stmt1)
# create output and input/output link messages to process
dbconAg.add(messageURI, RDF.TYPE, messageTypeURI)
dbconAg.add(messageURI, index, messageindexInt)
dbconAg.add(processURI, hasOutputRel, messageURI)
dbconAg.add(processURI, isRunOfProtocl, protocolTypeURI)
dbconAg.add(processURI, hasInputRel, prevMessageURI)
# prevMessageURI=messageURI
addedProcesses += 1
processes_ids.append(processid)
print "Request 7", str(id2)
return processes_ids
except Exception as e:
print e
return 404
def post(self, id, id1):
"""Add workflow
This method allows adding a workflow to a pipeline.
Requires workflow id and its location in the pipeline
Parameters
----------
id: str
project identifier
id1: str
pipeline identifier
Returns
-------
code: 201 if successfully added.
"""
# Agraph
args = pipeline_post_parser.parse_args()
wkflid = args.workflow_id
prtjctid = id
pplid = id1
step = args.step
wkflid = wkflid.split(',')
step = step.split(',')
# check if workflow is on pipeline
pipelineStr = localNSpace+"projects/"+str(prtjctid)+"/pipelines/"+\
str(pplid)
queryString = "SELECT ?execStep (STR(?intstep) as ?step) WHERE {<" + pipelineStr + "> obo:NGS_0000076 ?execStep. ?execStep obo:NGS_0000081 ?intstep.}"
tupleQuery = dbconAg.prepareTupleQuery(QueryLanguage.SPARQL,
queryString)
result = tupleQuery.evaluate()
jsonResult = parseAgraphQueryRes(result, ["execStep", "step"])
result.close()
for result in jsonResult:
aux1 = result["execStep"]
aux2 = result["step"]
step_converted = map(int, step)
if int(aux2.replace('"', '')) in step_converted \
or int(aux2.replace('"', '')) > max(step_converted):
toremove = dbconAg.createURI(aux1)
dbconAg.remove(None, None, toremove)
dbconAg.remove(toremove, None, None)
counter = -1
for i in wkflid:
counter += 1
# add new workflow
exStepType = dbconAg.createURI(namespace=obo,
localname="NGS_0000074")
workflowURI = dbconAg.createURI(namespace=localNSpace,
localname="workflows/" + str(i))
executeRel = dbconAg.createURI(namespace=obo,
localname="NGS_0000076")
pipelineURI = dbconAg.createURI(
namespace=localNSpace + "projects/",
localname=str(prtjctid) + "/pipelines/" + str(pplid))
exStepURI = dbconAg.createURI(namespace=localNSpace + "projects/",
localname=str(prtjctid) +
"/pipelines/" + str(pplid) +
"/step/" + str(step[counter]))
indexInt = dbconAg.createLiteral((step[counter]),
datatype=XMLSchema.INT)
indexProp = dbconAg.createURI(namespace=obo,
localname="NGS_0000081")
hasWorkflRel = dbconAg.createURI(namespace=obo,
localname="NGS_0000079")
dbconAg.add(exStepURI, RDF.TYPE, exStepType)
stmt1 = dbconAg.createStatement(exStepURI, indexProp, indexInt)
dbconAg.add(stmt1)
# link pipeline to step
dbconAg.add(pipelineURI, executeRel, exStepURI)
# add workflow + link to step
workflowType = dbconAg.createURI(namespace=obo,
localname="OBI_0500000")
dbconAg.add(workflowURI, RDF.TYPE, workflowType)
dbconAg.add(exStepURI, hasWorkflRel, workflowURI)
return 201
def set_process_output(project_id, pipeline_id, process_id, run_info,
run_stats, output, log_file, status):
try:
# Agraph
processURI = dbconAg.createURI(
namespace=localNSpace + "projects/",
localname=str(project_id) + "/pipelines/" + str(
pipeline_id) + "/processes/" + str(process_id))
# get output URI from process
hasOutput = dbconAg.createURI(namespace=obo,
localname="RO_0002234")
statements = dbconAg.getStatements(processURI, hasOutput, None)
outputURI = parseAgraphStatementsRes(statements)
statements.close()
outputURI = dbconAg.createURI(outputURI[0]['obj'])
runInfo = dbconAg.createLiteral((run_info),
datatype=XMLSchema.STRING)
runInfoProp = dbconAg.createURI(namespace=obo,
localname="NGS_0000092")
runStats = dbconAg.createLiteral((run_stats),
datatype=XMLSchema.STRING)
runStatsProp = dbconAg.createURI(namespace=obo,
localname="NGS_0000093")
runFile = dbconAg.createLiteral((output),
datatype=XMLSchema.STRING)
runFileProp = dbconAg.createURI(namespace=obo,
localname="NGS_0000094")
logFile = dbconAg.createLiteral((log_file),
datatype=XMLSchema.STRING)
logFileProp = dbconAg.createURI(namespace=obo,
localname="NGS_0000096")
runStatus = dbconAg.createLiteral((status),
datatype=XMLSchema.STRING)
runStatusProp = dbconAg.createURI(namespace=obo,
localname="NGS_0000097")
dbconAg.remove(outputURI, runInfoProp, None)
dbconAg.remove(outputURI, runStatsProp, None)
dbconAg.remove(outputURI, runFileProp, None)
dbconAg.remove(outputURI, runStatusProp, None)
# add outputs paths to process
stmt1 = dbconAg.createStatement(outputURI, runInfoProp,
runInfo)
stmt2 = dbconAg.createStatement(outputURI, runStatsProp,
runStats)
stmt3 = dbconAg.createStatement(outputURI, runFileProp,
runFile)
stmt4 = dbconAg.createStatement(outputURI, logFileProp, logFile)
stmt5 = dbconAg.createStatement(outputURI, runStatusProp,
runStatus)
# send to allegro
dbconAg.add(stmt1)
dbconAg.add(stmt2)
dbconAg.add(stmt3)
dbconAg.add(stmt4)
dbconAg.add(stmt5)
except Exception as e:
print "ERROR", e