def installed_tool_list(self):
"""
gets a tool list from the toolclient
:return:
"""
tool_client = ToolClient(self.gi)
return tool_client.get_tools()
def main():
try:
input_path = sys.argv[1]
output_path = sys.argv[2]
galaxyInstance = GalaxyInstance(url = GALAXY_URL, key=API_KEY)
historyClient = HistoryClient(galaxyInstance)
toolClient = ToolClient(galaxyInstance)
workflowClient = WorkflowClient(galaxyInstance)
datasetClient = DatasetClient(galaxyInstance)
history = historyClient.create_history('tmp')
uploadedFile = toolClient.upload_file(input_path, history['id'] )
workflow = workflowClient.show_workflow(WORKFLOW_ID)
dataset_map = {workflow['inputs'].keys()[0]: {'id': uploadedFile['outputs'][0]['id'], 'src': 'hda'}}
params = {TOOL_ID_IN_GALAXY: {'param': 'reference_genome', 'value': 'hg19'}}
output = workflowClient.run_workflow(WORKFLOW_ID, dataset_map, params, history['id'])
downloadDataset(datasetClient, findDatasedIdByExtention(datasetClient, output, 'bed'), output_path)
#delete history
historyClient.delete_history(history['id'])
#if galaxy instance support dataset purging
#historyClient.delete_history(history['id'], True)
except IndexError:
print 'usage: %s key url workflow_id history step=src=dataset_id' % os.path.basename(sys.argv[0])
sys.exit(1)
def installed_tool_list(self):
"""
gets a tool list from the toolclient
:return:
"""
tool_client = ToolClient(self.gi)
return tool_client.get_tools()
def put(filename, file_type = 'auto'):
"""
Given a filename of any file accessible to the docker instance, this
function will upload that file to galaxy using the current history.
Does not return anything.
"""
conf = _get_conf()
gi = get_galaxy_connection()
tc = ToolClient( gi )
tc.upload_file(filename, conf['history_id'], file_type = file_type)
def put(filename, file_type='auto', history_id=None):
"""
Given a filename of any file accessible to the docker instance, this
function will upload that file to galaxy using the current history.
Does not return anything.
"""
conf = _get_conf()
gi = get_galaxy_connection()
tc = ToolClient(gi)
history_id = history_id or _get_history_id()
tc.upload_file(filename, history_id, file_type=file_type)
def transfer_filelist_from_ftp(gi, filelist, history_name):
tc = ToolClient(gi)
hc = HistoryClient(gi)
st = get_time_stamp()
hist = hc.create_history('{}-{}'.format(history_name, st))
uploaded_files = []
for f in filelist:
upf = tc.upload_from_ftp(path=os.path.basename(f),
history_id=hist['id'])['outputs'][0]
print(upf)
uploaded_files.append(upf)
return uploaded_files, hist
def create_clients(self):
'''
Create bioblend clients for the Galaxy instance.
'''
# Create first client and check if the API works
self.config_client = ConfigClient(self.instance)
try:
self.config_client.get_version()
self.config_client.get_config()
except:
logger.error("Provided API-key does not work.")
return False
try:
self.user_client = UserClient(self.instance)
self.workflow_client = WorkflowClient(self.instance)
self.tool_client = ToolClient(self.instance)
self.toolshed_client = ToolShedClient(self.instance)
self.library_client = LibraryClient(self.instance)
self.roles_client = RolesClient(self.instance)
self.history_client = HistoryClient(self.instance)
self.dataset_client = DatasetClient(self.instance)
except:
logger.error("Error initializing other bioblend clients.")
return False
return True
def main():
galaxyInstance = GalaxyInstance(url=GALAXY_URL, key=API_KEY)
toolClient = ToolClient(galaxyInstance)
histories = HistoryClient(galaxyInstance)
workflowsClient = WorkflowClient(galaxyInstance)
libraryClient = LibraryClient(galaxyInstance)
brassica_library = libraryClient.get_libraries(
name=' Evolutionary Systems Biology')
files = libraryClient.show_library(brassica_library[0]['id'],
contents=True)
#print(files)
itemp = 0
for f in files:
if f['type'] == 'folder':
continue # do nothing, try next
#initial set
#if itemp == 31:
# break
#print ("Name " + f['name'])
replicate = f['name'].split('_')[-1].split('.')[0]
#print replicate
if replicate == '1':
itemp = itemp + 1
if not (itemp >= 71 and itemp <= 92):
continue
base = f['name'].split('_')[:-1]
#print base
forward_name = f['name']
reverse_name = '_'.join(base) + '_2.fastq.bz2'
forward_id = f['id']
files2 = libraryClient.show_library(brassica_library[0]['id'],
contents=True)
for f2 in files2:
if f2['name'] == reverse_name:
reverse_id = f2['id']
print forward_name
print reverse_name
new_history_name = f['name'].split('_')[7] + "_" + f['name'].split(
'_')[-3] + "_" + f['name'].split('_')[-2]
print new_history_name
hist = histories.create_history(name=new_history_name)
dataset_F = histories.upload_dataset_from_library(
hist['id'], forward_id)
dataset_R = histories.upload_dataset_from_library(
hist['id'], reverse_id)
datamap = {}
datamap['0'] = {'src': 'hda', 'id': dataset_F['id']}
datamap['1'] = {'src': 'hda', 'id': dataset_R['id']}
workflows = workflowsClient.get_workflows(name="Maize HISAT 2.1")
workflow = workflows[0]
try:
w = workflowsClient.run_workflow(workflow['id'],
datamap,
history_id=hist['id'])
except:
print('Next')
def main():
"""
This script uses bioblend to trigger dependencies installations for the provided tools
"""
args = _parser().parse_args()
gi = get_galaxy_connection(args)
tool_client = ToolClient(gi)
if args.verbose:
log.basicConfig(level=log.DEBUG)
if args.tool:
for tool_conf_path in args.tool: # type: str
_, ext = os.path.splitext(tool_conf_path)
if ext == '.xml':
log.info("tool_conf xml found, parsing..")
# install all
root = ET.ElementTree(file=tool_conf_path).getroot()
if root.tag == "toolbox":
# Install all from tool_conf
tool_path = root.get('tool_path', '')
tool_path = tool_path.replace(
'${tool_conf_dir}',
os.path.abspath(os.path.dirname(tool_conf_path)))
if tool_path:
log.info("Searching for tools relative to " +
tool_path)
tools = root.findall(".//tool[@file]")
if len(tools) == 0:
log.warning("No tools found in tool_conf")
continue
for tool in tools:
tool_id = ET.ElementTree(file=os.path.join(
tool_path, tool.get('file'))).getroot().get('id')
if tool_id:
log.info("Installing tool dependencies for " +
tool_id + " from: " + tool.get('file'))
_install(tool_client, tool_id)
elif root.tag == "tool" and root.get('id'):
# Install from single tool file
log.info("Tool xml found. Installing " + root.get('id') +
" dependencies..")
_install(tool_client, root.get('id'))
else:
log.info("YAML tool list found, parsing..")
with open(tool_conf_path) as fh:
tool_ids = yaml.safe_load(fh)
for tool_id in tool_ids:
# Install from yaml file
log.info("Installing " + tool_id + " dependencies..")
_install(tool_client, tool_id)
if args.id:
for tool_id in args.id: # type: str
log.info("Installing " + tool_id + " dependencies..")
_install(tool_client, tool_id.strip())
def put(filename,
file_type='auto',
history_id=None,
use_objects=DEFAULT_USE_OBJECTS):
"""
Given a filename of any file accessible to the docker instance, this
function will upload that file to galaxy using the current history.
Does not return anything.
"""
conf = _get_conf()
gi = get_galaxy_connection(use_objects)
history_id = history_id or _get_history_id()
if use_objects:
history = gi.histories.get(history_id)
history.upload_dataset(filename, file_type=file_type)
else:
tc = ToolClient(gi)
tc.upload_file(filename, history_id, file_type=file_type)
def __init__(self, galaxy_instance, configuration):
"""
:param galaxy_instance: A GalaxyInstance object (import from bioblend.galaxy)
:param configuration: A dictionary. Examples in the ephemeris documentation.
"""
self.gi = galaxy_instance
self.config = configuration
self.tool_data_client = ToolDataClient(self.gi)
self.tool_client = ToolClient(self.gi)
self.possible_name_keys = ['name', 'sequence_name'] # In order of importance!
self.possible_value_keys = ['value', 'sequence_id', 'dbkey'] # In order of importance!
self.data_managers = self.config.get('data_managers')
self.genomes = self.config.get('genomes', '')
self.source_tables = DEFAULT_SOURCE_TABLES
self.fetch_jobs = []
self.skipped_fetch_jobs = []
self.index_jobs = []
self.skipped_index_jobs = []
def main():
galaxyInstance = GalaxyInstance(url=GALAXY_URL, key=API_KEY)
toolClient = ToolClient(galaxyInstance)
histories = HistoryClient(galaxyInstance)
workflowsClient = WorkflowClient(galaxyInstance)
libraryClient = LibraryClient(galaxyInstance)
brassica_library = libraryClient.get_libraries(
name=' Evolutionary Systems Biology')
files = libraryClient.show_library(brassica_library[0]['id'],
contents=True)
#print(files)
for f in files:
if f['type'] == 'folder':
continue # do nothing, try next
#initial set
#if itemp == 31:
# break
#print ("Name " + f['name'])
replicate = f['name'].split('/')[-1][0]
#print replicate
if replicate == 'X':
base = f['name'].split('/')[-1].split('.')[0]
#print base
forward_name = f['name']
forward_id = f['id']
print forward_name
new_history_name = base
print new_history_name
hist = histories.create_history(name=new_history_name)
dataset_F = histories.upload_dataset_from_library(
hist['id'], forward_id)
datamap = {}
datamap['0'] = {'src': 'hda', 'id': dataset_F['id']}
workflows = workflowsClient.get_workflows(
name="Maize Small samples HISAT 2.1")
workflow = workflows[0]
try:
w = workflowsClient.run_workflow(workflow['id'],
datamap,
history_id=hist['id'])
except:
print('Next')
def main():
galaxyInstance = GalaxyInstance(url=GALAXY_URL, key=API_KEY)
toolClient = ToolClient(galaxyInstance)
historyClient = HistoryClient(galaxyInstance)
workflowsClient = WorkflowClient(galaxyInstance)
libraryClient = LibraryClient(galaxyInstance)
datasetClient = DatasetClient(galaxyInstance)
histories = historyClient.get_histories(deleted=False)
for hist in histories:
hist_id = hist['id']
countSecondary = historyClient.show_matching_datasets(
hist_id, name_filter=name_filter)
if len(countSecondary) != 0:
#print(countSecondary)
file_path = dir_name + '/' + hist[
'name'] + '_' + name_filter + '.' + ext
#print(file_path)
#print(countSecondary[0]['dataset_id'])
datasetClient.download_dataset(countSecondary[0]['id'],
file_path=file_path,
use_default_filename=False)
sys.exit()
def runWorkflow(argDictionary, comparisons,samples):
from bioblend.galaxy import GalaxyInstance
from bioblend.galaxy.histories import HistoryClient
from bioblend.galaxy.tools import ToolClient
from bioblend.galaxy.workflows import WorkflowClient
from bioblend.galaxy.libraries import LibraryClient
import tempfile
import time
api_key = ''
galaxy_host = ''
gi = GalaxyInstance(url=galaxy_host, key=api_key)
history_client = HistoryClient(gi)
tool_client = ToolClient(gi)
workflow_client = WorkflowClient(gi)
library_client = LibraryClient(gi)
history = history_client.create_history(argDictionary['accessionNumber'])
comparisonsTable = tool_client.upload_file(comparisons, history['id'], file_type='txt')
sampleTable = tool_client.upload_file(samples, history['id'], file_type='tabular')
if argDictionary['site'] == "ENA":
#fastqs available on ENA
tool_inputs = {
"accessionNumber":argDictionary["ENA"],"sampleTable":{'id': sampleTable['outputs'][0]['id'], 'src': 'hda'}
}
#run the tool to get the data from ENA
tool_client.run_tool(history['id'],'getRNASeqExpressionData', tool_inputs)
#we want to wait until we have all datasets
while getNumberNotComplete(history['id'], history_client) > 0:
time.sleep(10)
#sleep until all the fastq files are findable
time.sleep(120)
dirpath = tempfile.mkdtemp()
fileList = getDatasetsByApproxName("files.tabular", history,history_client)[0]
fileList = history_client.download_dataset(history["id"],fileList["id"],dirpath)
num_lines = sum(1 for line in open(fileList)) -1
datasets=list()
while len(datasets)!=num_lines:
time.sleep(10)
datasets = getDatasetsByApproxName("fastq",history,history_client )
else: #for SRA
if argDictionary['single'] == "TRUE":
with open(samples) as tsvfile:
reader = csv.DictReader(tsvfile, delimiter='\t')
for sample in reader:
print (sample)
fileNames=str.split(sample["File"],"|")
for fileName in fileNames:
tool_inputs = {
"input|input_select":"accession_number",
"outputformat":"fastqsanger.gz",
"input|accession":fileName
}
#run the tool to get the single data from SRA
tool_client.run_tool(history['id'],'toolshed.g2.bx.psu.edu/repos/iuc/sra_tools/fastq_dump/2.8.1.3', tool_inputs)
else:
with open(samples) as tsvfile:
reader = csv.DictReader(tsvfile, delimiter='\t')
for sample in reader:
tool_inputs = {
"accession_number":sample["File"]
}
#run the tool to get the paired data from SRA
tool_client.run_tool(history['id'],'toolshed.g2.bx.psu.edu/repos/mandorodriguez/fastqdump_paired/fastq_dump_paired/1.1.4', tool_inputs)
while getNumberNotComplete(history['id'], history_client) > 0:
time.sleep(10)
datasets = getDatasetsByApproxName("fastq",history,history_client )
#get the fastQC tool
for fastq in datasets:
try:
tool_inputs = {'input_file' : {'id': fastq['id'], 'src': 'hda'}}
tool_client.run_tool(history['id'],'toolshed.g2.bx.psu.edu/repos/devteam/fastqc/fastqc/0.69', tool_inputs)
except Exception:
pass
#wait till complete
while getNumberNotComplete(history['id'], history_client) > 0:
time.sleep(10)
#make dataset collections for quantification using the fastq files
collections=list()
with open(samples) as tsvfile:
reader = csv.DictReader(tsvfile, delimiter='\t')
for row in reader:
datasets=list()
fileNames=str.split(row["File"],"|")
for fileName in fileNames:
datasets= datasets + getDatasetsByApproxName(fileName,history,history_client )
#make list of datasets
collections.append(makeDataSetCollection(datasets,row["Sample"],history,history_client))
#get the correct kallisto index
species = argDictionary['species'].lower()
index = getLibraryToolDataID(history=history,history_client=history_client,library_client=library_client,name=species +"IndexFile")
index = {'id': index, 'src': 'hda'}
#run kallisto for every dataset collection
for collection in collections:
#set up the tool_inputs
tool_inputs = {'index' : index,'inputs' : {'id': collection['id'], 'src': 'hdca'} ,"single":argDictionary["single"],"stranded":argDictionary["stranded"]}
#often encounter connection broken error - possible problem with Certus server?
#bypass by ignoring the exception
tool_client.run_tool(history['id'],'kallistoQuant', tool_inputs)
# we want to wait until we have all datasets
while getNumberNotComplete(history['id'], history_client) > 0:
time.sleep(10)
# Run multiqc on kallisto logs and fastqc files
datasets = getDatasetsByApproxName("RawData",history,history_client )
kallistoLogs = getDatasetsByApproxName(".log", history, history_client)
tool_inputs = {}
for i, dataset in enumerate(datasets+kallistoLogs):
if not dataset["deleted"]:
if dataset in datasets:
software = 'fastqc'
else:
software = 'kallisto'
params = {'id' : dataset['id'], 'src': 'hda', 'name': dataset['name']}
tool_inputs.update({'results_%s|software_cond|software' % i: software, 'results_%s|input_file' % i: params})
# #summarise with the multiQC tool
tool_client.run_tool(history['id'],'multiqc', tool_inputs)
multiQc = getDatasetsByApproxName("multiqc",history,history_client)[0]
#get all the abundance files to convert to gene level counts matrix
datasets = getDatasetsByApproxName(".abundance",history,history_client )
#make a dataset collection for to make a countsMatrix
collection = makeDataSetCollection(datasets,"abundances",history,history_client)
#set up the tool_inputs
tool_inputs = {'inputs' : {'id': collection['id'], 'src': 'hdca'} ,"species":argDictionary['species']}
#convert abundances to gene level counts matrix
tool_client.run_tool(history['id'],'KallistoAbundancestoGeneCountMatrix', tool_inputs)
# A diry hack, we want to wait until we have all datasets
while getNumberNotComplete(history['id'], history_client) > 0:
time.sleep(10)
txi = getDatasetsByApproxName("txi",history,history_client)
#set up the tool_inputs for PCA
tool_inputs = {'txiData' : {'id': txi[0]['id'], 'src': 'hda'} ,'sampleTable' : {'id': sampleTable['outputs'][0]['id'], 'src': 'hda'} ,"species":argDictionary['species'],'technicalReplicates':argDictionary['technicalReplicates'],'batchCorrect':argDictionary['batchCorrect']}
#run deseq2
tool_client.run_tool(history['id'],'PCARNASeq', tool_inputs)
pca = getDatasetsByApproxName("PCA",history,history_client)[0]
#set up the tool_inputs for DESeq2
tool_inputs = {'txiData' : {'id': txi[0]['id'], 'src': 'hda'} ,'sampleTable' : {'id': sampleTable['outputs'][0]['id'], 'src': 'hda'} ,
'comparisonsTable' : {'id': comparisonsTable['outputs'][0]['id'], 'src': 'hda'} ,"foldChangeOnly":argDictionary['foldChangeOnly'],"species":argDictionary['species'],'technicalReplicates':argDictionary['technicalReplicates'],'batchCorrect':argDictionary['batchCorrect']}
#run deseq2
tool_client.run_tool(history['id'],'DESeq2FoldChange', tool_inputs)
#run chrdir
tool_client.run_tool(history['id'],'characteristicDirectionRNASeq', tool_inputs)
#we want to wait until we have all datasets
while getNumberNotComplete(history['id'], history_client) > 0:
time.sleep(10)
#get the foldchange data, cut and run pathway workflow
dataset_id = getFoldChangeData(history, history_client)['id']
return_collection = [{'accessionNo':argDictionary['accessionNumber'], 'foldChange': getUrl(dataset_id), 'PCA': getUrl(pca["id"]),'chrDirTable': getUrl(getMostRecentDatasetByName('chrDirTable.tabular', history, history_client)['id'])}]
number_of_comparisons = -1
for line in open(comparisons):
if not line.isspace():
number_of_comparisons += 1
for comparison in range(0, int(number_of_comparisons)):
tool_inputs = {
'foldChangeTable' : {'id': dataset_id, 'src': 'hda'},
'comparisonNumber' : comparison + 1
}
tool_client.run_tool(history['id'], 'cutFoldChangeTable', tool_inputs)
while getNumberNotComplete(history['id'], history_client) > 0:
time.sleep(10)
if argDictionary['species'] in ["Rat","Cow","Horse","Pig","Zebrafish"]:
pathwayAnalysisWorkflow = workflow_client.show_workflow('c9468fdb6dc5c5f1')
params = dict()
for key in pathwayAnalysisWorkflow['steps'].keys():
params[key] = argDictionary
if argDictionary['species'] == "Rat":
network=getLibraryToolDataID(history=history,history_client=history_client,library_client=library_client,name="ratStringNetwork")
geneLengths=getLibraryToolDataID(history=history,history_client=history_client,library_client=library_client,name="ratGeneLengths")
homology=getLibraryToolDataID(history=history,history_client=history_client,library_client=library_client,name="HOM_AllOrganism.rpt")
if argDictionary['species'] == "Cow":
network=getLibraryToolDataID(history=history,history_client=history_client,library_client=library_client,name="cowStringNetwork")
geneLengths=getLibraryToolDataID(history=history,history_client=history_client,library_client=library_client,name="cowGeneLengths")
homology=getLibraryToolDataID(history=history,history_client=history_client,library_client=library_client,name="HOM_AllOrganism.rpt")
if argDictionary['species'] == "Horse":
network=getLibraryToolDataID(history=history,history_client=history_client,library_client=library_client,name="horseStringNetwork")
geneLengths=getLibraryToolDataID(history=history,history_client=history_client,library_client=library_client,name="horseGeneLengths")
homology=getLibraryToolDataID(history=history,history_client=history_client,library_client=library_client,name="Homology.horse.txt")
if argDictionary['species'] == "Pig":
network=getLibraryToolDataID(history=history,history_client=history_client,library_client=library_client,name="pigStringNetwork.txt")
geneLengths=getLibraryToolDataID(history=history,history_client=history_client,library_client=library_client,name="pigGeneLengths.tabular")
homology=getLibraryToolDataID(history=history,history_client=history_client,library_client=library_client,name="Homology.pig.txt")
if argDictionary['species'] == "Zebrafish":
network=getLibraryToolDataID(history=history,history_client=history_client,library_client=library_client,name="zebrafishStringNetwork")
geneLengths=getLibraryToolDataID(history=history,history_client=history_client,library_client=library_client,name="zebrafishGeneLengths")
homology=getLibraryToolDataID(history=history,history_client=history_client,library_client=library_client,name="HOM_AllOrganism.rpt")
pathwayDatamap = {'3' : {'id': homology, 'src': 'hda'},'2' : {'id': network, 'src': 'hda'},'1' : {'id': geneLengths, 'src': 'hda'}}
diffExpDataCollection = getDatasetsByName('cutTable.tabular', history, history_client)
for index, diffExpData in enumerate(diffExpDataCollection):
numCompleted = getNumberComplete(history['id'], history_client) + 10
print(numCompleted)
pathwayDatamap["0"] = {'id': diffExpData['id'], 'src': 'hda'}
workflow_client.invoke_workflow(pathwayAnalysisWorkflow['id'],
inputs = pathwayDatamap,
history_id = history['id'],
params = params)
comparisonDict = getRowFromCsv(comparisons, index)
if 'Factor1' in comparisonDict.keys():
comparisonDict['Factor'] = comparisonDict['Factor1'] + "." + comparisonDict['Factor2']
return_dict = {'accessionNo':argDictionary['accessionNumber'],
'factor':comparisonDict['Factor'],
'comparisonNum':comparisonDict['Numerator'],
'comparisonDenom':comparisonDict['Denominator'],
'foldChange': getUrl(diffExpData['id']),
'interactome': pathwayDatamap['0']['id'],
'exonLength': pathwayDatamap['2']['id']}
while getNumberComplete(history['id'], history_client) < numCompleted:
time.sleep(10)
return_dict['moduleNodes'] = getUrl(getMostRecentDatasetByName('moduleNodes.text',
history, history_client)['id'])
return_dict['modulePlots'] = getUrl(getMostRecentDatasetByName('modulePlots.pdf',
history, history_client)['id'])
return_dict['slimEnrichPathways'] = getUrl(getMostRecentDatasetByName('slimEnrichmentPathways.tabular',
history, history_client)['id'])
return_dict['enrichedDrugsReverse'] = getUrl(getMostRecentDatasetByName('enrichedDrugsReverse.tabular',
history, history_client)['id'])
return_dict['enrichedDrugsMimic'] = getUrl(getMostRecentDatasetByName('enrichedDrugsMimic.tabular',
history, history_client)['id'])
return_dict['enrichedTerms'] = getUrl(getMostRecentDatasetByName('enrichedTerms.tabular',
history, history_client)['id'])
return_dict['enrichedTerms.reduced'] = getUrl(getMostRecentDatasetByName('enrichedTerms.reduced.tabular',
history, history_client)['id'])
return_dict['GO.MDS'] = getUrl(getMostRecentDatasetByName('GO.MDS.html',
history, history_client)['id'])
return_collection.append(return_dict)
# Hard code keys to define the order
keys = ['accessionNo','multiQC','factor','PCA','chrDirTable','comparisonNum','comparisonDenom','foldChange',
'interactome','exonLength','moduleNodes','modulePlots',
'slimEnrichPathways','secretedProteins','enrichedDrugsReverse','enrichedDrugsMimic','enrichedTerms','enrichedTerms.reduced','GO.MDS']
outFileName = 'output/' + argDictionary['accessionNumber'] + '-workflowOutput.tsv'
with open(outFileName, 'wb') as csvFile:
# Get headers from last dictionary in collection as first doesn't contain all keys
csvOutput = csv.DictWriter(csvFile, keys, delimiter = "\t")
csvOutput.writeheader()
csvOutput.writerows(return_collection)
#tool_client.upload_file(outFileName, history['id'], file_type='tsv')
return return_collection
else:
pathwayAnalysisWorkflow = workflow_client.show_workflow('e85a3be143d5905b')
params = dict()
for key in pathwayAnalysisWorkflow['steps'].keys():
params[key] = argDictionary
if argDictionary['species'] == "Mouse":
network=getLibraryToolDataID(history=history,history_client=history_client,library_client=library_client,name="mouseStringNetwork")
geneLengths=getLibraryToolDataID(history=history,history_client=history_client,library_client=library_client,name="MouseGeneLengths.tab")
homology=getLibraryToolDataID(history=history,history_client=history_client,library_client=library_client,name="Homology.mouse.txt")
secretedReference=getLibraryToolDataID(history=history,history_client=history_client,library_client=library_client,name="uniprot-secreted-mouse.txt")
pathwayDatamap = {'4' : {'id': secretedReference, 'src': 'hda'},'3' : {'id': homology, 'src': 'hda'},'2' : {'id': network, 'src': 'hda'},'1' : {'id': geneLengths, 'src': 'hda'}}
else:
network=getLibraryToolDataID(history=history,history_client=history_client,library_client=library_client,name="humanStringNetwork")
geneLengths=getLibraryToolDataID(history=history,history_client=history_client,library_client=library_client,name="geneLengths")
homology=getLibraryToolDataID(history=history,history_client=history_client,library_client=library_client,name="Homology.mouse.txt")
secretedReference=getLibraryToolDataID(history=history,history_client=history_client,library_client=library_client,name="uniprot-secreted-human.txt")
pathwayDatamap = {'4' : {'id': secretedReference, 'src': 'hda'},'3' : {'id': homology, 'src': 'hda'},'2' : {'id': network, 'src': 'hda'},'1' : {'id': geneLengths, 'src': 'hda'}}
diffExpDataCollection = getDatasetsByName('cutTable.tabular', history, history_client)
for index, diffExpData in enumerate(diffExpDataCollection):
numCompleted = getNumberComplete(history['id'], history_client) + 14
print(numCompleted)
pathwayDatamap["0"] = {'id': diffExpData['id'], 'src': 'hda'}
#pathwayDatamap['1'] = {'id': diffExpData['id'], 'src': 'hda'}
workflow_client.invoke_workflow(pathwayAnalysisWorkflow['id'],
inputs = pathwayDatamap,
history_id = history['id'],
params = params)
comparisonDict = getRowFromCsv(comparisons, index)
if 'Factor1' in comparisonDict.keys():
comparisonDict['Factor'] = comparisonDict['Factor1'] + "." + comparisonDict['Factor2']
return_dict = {'accessionNo':argDictionary['accessionNumber'],
'factor':comparisonDict['Factor'],
'comparisonNum':comparisonDict['Numerator'],
'comparisonDenom':comparisonDict['Denominator'],
'foldChange': getUrl(diffExpData['id']),
'interactome': pathwayDatamap['0']['id'],
'exonLength': pathwayDatamap['2']['id']}
while getNumberComplete(history['id'], history_client) < numCompleted:
time.sleep(10)
return_dict['moduleNodes'] = getUrl(getMostRecentDatasetByName('moduleNodes.text',
history, history_client)['id'])
return_dict['modulePlots'] = getUrl(getMostRecentDatasetByName('modulePlots.pdf',
history, history_client)['id'])
return_dict['pathways'] = getUrl(getMostRecentDatasetByName('pathways.tabular',
history, history_client)['id'])
return_dict['enrichPlot'] = getUrl(getMostRecentDatasetByName('enrichmentPlot.png',
history, history_client)['id'])
return_dict['enrichmentTable'] = getUrl(getMostRecentDatasetByName('TF_EnrichmentTable.tabular',
history, history_client)['id'])
return_dict['slimEnrichPathways'] = getUrl(getMostRecentDatasetByName('slimEnrichmentPathways.tabular',
history, history_client)['id'])
return_dict['secretedProteins'] = getUrl(getMostRecentDatasetByName('secretedProteins.tabular',
history, history_client)['id'])
return_dict['enrichedDrugsReverse'] = getUrl(getMostRecentDatasetByName('enrichedDrugsReverse.tabular',
history, history_client)['id'])
return_dict['enrichedDrugsMimic'] = getUrl(getMostRecentDatasetByName('enrichedDrugsMimic.tabular',
history, history_client)['id'])
return_dict['enrichedTerms'] = getUrl(getMostRecentDatasetByName('enrichedTerms.tabular',
history, history_client)['id'])
return_dict['enrichedTerms.reduced'] = getUrl(getMostRecentDatasetByName('enrichedTerms.reduced.tabular',
history, history_client)['id'])
return_dict['GO.MDS'] = getUrl(getMostRecentDatasetByName('GO.MDS.html',
history, history_client)['id'])
return_collection.append(return_dict)
# Hard code keys to define the order
keys = ['accessionNo','multiQC','factor','PCA','chrDirTable','comparisonNum','comparisonDenom','foldChange',
'interactome','exonLength','moduleNodes','modulePlots','pathways','enrichPlot', 'enrichmentTable',
'slimEnrichPathways','secretedProteins','enrichedDrugsReverse','enrichedDrugsMimic','enrichedTerms','enrichedTerms.reduced','GO.MDS']
outFileName = 'output/' + argDictionary['accessionNumber'] + '-workflowOutput.tsv'
with open(outFileName, 'wb') as csvFile:
# Get headers from last dictionary in collection as first doesn't contain all keys
csvOutput = csv.DictWriter(csvFile, keys, delimiter = "\t")
csvOutput.writeheader()
csvOutput.writerows(return_collection)
return return_collection
#!/usr/bin/python
# version 1.0.0 18-Apr-2018
"""
This tools imports a data set into galaxy
"""
import sys
###REPLACE### sys.path.append('/PATH/TO/lib/python2.7/site-packages/bioblend-0.8.0-py2.7.egg')
###REPLACE### sys.path.append('/PATH/TOlib/python2.7/site-packages/requests_toolbelt-0.7.0-py2.7.egg')
from bioblend.galaxy import GalaxyInstance
from bioblend.galaxy.tools import ToolClient
###REPLACE### url = "GALAXY URL"
###REPLACE### key = "API KEY of an ADMIN USER"
history_id = sys.argv[1]
file_path = sys.argv[2]
#print("Initiating Galaxy connection")
gi = GalaxyInstance(url=url, key=key)
toolClient = ToolClient(gi)
uploadedFile = toolClient.upload_file(file_path, history_id)
class DataManagers:
def __init__(self, galaxy_instance, configuration):
"""
:param galaxy_instance: A GalaxyInstance object (import from bioblend.galaxy)
:param configuration: A dictionary. Examples in the ephemeris documentation.
"""
self.gi = galaxy_instance
self.config = configuration
self.tool_data_client = ToolDataClient(self.gi)
self.tool_client = ToolClient(self.gi)
self.possible_name_keys = ['name',
'sequence_name'] # In order of importance!
self.possible_value_keys = ['value', 'sequence_id',
'dbkey'] # In order of importance!
self.data_managers = self.config.get('data_managers')
self.genomes = self.config.get('genomes', '')
self.source_tables = DEFAULT_SOURCE_TABLES
self.fetch_jobs = []
self.skipped_fetch_jobs = []
self.index_jobs = []
self.skipped_index_jobs = []
def initiate_job_lists(self):
"""
Determines which data managers should be run to populate the data tables.
Distinguishes between fetch jobs (download files) and index jobs.
:return: populate self.fetch_jobs, self.skipped_fetch_jobs, self.index_jobs and self.skipped_index_jobs
"""
self.fetch_jobs = []
self.skipped_fetch_jobs = []
self.index_jobs = []
self.skipped_index_jobs = []
for dm in self.data_managers:
jobs, skipped_jobs = self.get_dm_jobs(dm)
if self.dm_is_fetcher(dm):
self.fetch_jobs.extend(jobs)
self.skipped_fetch_jobs.extend(skipped_jobs)
else:
self.index_jobs.extend(jobs)
self.skipped_index_jobs.extend(skipped_jobs)
def get_dm_jobs(self, dm):
"""Gets the job entries for a single dm. Puts entries that already present in skipped_job_list.
:returns job_list, skipped_job_list"""
job_list = []
skipped_job_list = []
items = self.parse_items(dm.get('items', ['']))
for item in items:
dm_id = dm['id']
params = dm['params']
inputs = dict()
# Iterate over all parameters, replace occurences of {{item}} with the current processing item
# and create the tool_inputs dict for running the data manager job
for param in params:
key, value = list(param.items())[0]
value_template = Template(value)
value = value_template.render(item=item)
inputs.update({key: value})
job = dict(tool_id=dm_id, inputs=inputs)
data_tables = dm.get('data_table_reload', [])
if self.input_entries_exist_in_data_tables(data_tables, inputs):
skipped_job_list.append(job)
else:
job_list.append(job)
return job_list, skipped_job_list
def dm_is_fetcher(self, dm):
"""Checks whether the data manager fetches a sequence instead of indexing.
This is based on the source table.
:returns True if dm is a fetcher. False if it is not."""
data_tables = dm.get('data_table_reload', [])
for data_table in data_tables:
if data_table in self.source_tables:
return True
return False
def data_table_entry_exists(self, data_table_name, entry, column='value'):
"""Checks whether an entry exists in the a specified column in the data_table."""
try:
data_table_content = self.tool_data_client.show_data_table(
data_table_name)
except Exception:
raise Exception('Table "%s" does not exist' % data_table_name)
try:
column_index = data_table_content.get('columns').index(column)
except IndexError:
raise IndexError('Column "%s" does not exist in %s' %
(column, data_table_name))
for field in data_table_content.get('fields'):
if field[column_index] == entry:
return True
return False
def input_entries_exist_in_data_tables(self, data_tables, input_dict):
"""Checks whether name and value entries from the input are already present in the data tables.
If an entry is missing in of the tables, this function returns False"""
value_entry = get_first_valid_entry(input_dict,
self.possible_value_keys)
name_entry = get_first_valid_entry(input_dict, self.possible_name_keys)
# Return False if name and value entries are both None
if not value_entry and not name_entry:
return False
# Check every data table for existence of name and value
# Return False as soon as entry is not present
for data_table in data_tables:
if value_entry:
if not self.data_table_entry_exists(
data_table, value_entry, column='value'):
return False
if name_entry:
if not self.data_table_entry_exists(
data_table, name_entry, column='name'):
return False
# If all checks are passed the entries are present in the database tables.
return True
def parse_items(self, items):
"""
Parses items with jinja2.
:param items: the items to be parsed
:return: the parsed items
"""
if bool(self.genomes):
items_template = Template(json.dumps(items))
rendered_items = items_template.render(
genomes=json.dumps(self.genomes))
# Remove trailing " if present
rendered_items = rendered_items.strip('"')
items = json.loads(rendered_items)
return items
def run(self, log=None, ignore_errors=False, overwrite=False):
"""
Runs the data managers.
:param log: The log to be used.
:param ignore_errors: Ignore erroring data_managers. Continue regardless.
:param overwrite: Overwrite existing entries in data tables
"""
self.initiate_job_lists()
all_succesful_jobs = []
all_failed_jobs = []
all_skipped_jobs = []
if not log:
log = logging.getLogger()
def run_jobs(jobs, skipped_jobs):
job_list = []
for skipped_job in skipped_jobs:
if overwrite:
log.info(
'%s already run for %s. Entry will be overwritten.' %
(skipped_job["tool_id"], skipped_job["inputs"]))
jobs.append(skipped_job)
else:
log.info('%s already run for %s. Skipping.' %
(skipped_job["tool_id"], skipped_job["inputs"]))
all_skipped_jobs.append(skipped_job)
for job in jobs:
started_job = self.tool_client.run_tool(
history_id=None,
tool_id=job["tool_id"],
tool_inputs=job["inputs"])
log.info(
'Dispatched job %i. Running DM: "%s" with parameters: %s' %
(started_job['outputs'][0]['hid'], job["tool_id"],
job["inputs"]))
job_list.append(started_job)
successful_jobs, failed_jobs = wait(self.gi, job_list, log)
if failed_jobs:
if not ignore_errors:
log.error('Not all jobs successful! aborting...')
raise RuntimeError('Not all jobs successful! aborting...')
else:
log.warning('Not all jobs successful! ignoring...')
all_succesful_jobs.extend(successful_jobs)
all_failed_jobs.extend(failed_jobs)
log.info(
"Running data managers that populate the following source data tables: %s"
% self.source_tables)
run_jobs(self.fetch_jobs, self.skipped_fetch_jobs)
log.info("Running data managers that index sequences.")
run_jobs(self.index_jobs, self.skipped_index_jobs)
log.info('Finished running data managers. Results:')
log.info('Successful jobs: %i ' % len(all_succesful_jobs))
log.info('Skipped jobs: %i ' % len(all_skipped_jobs))
log.info('Failed jobs: %i ' % len(all_failed_jobs))
InstallResults = namedtuple(
"InstallResults",
["successful_jobs", "failed_jobs", "skipped_jobs"])
return InstallResults(successful_jobs=all_succesful_jobs,
failed_jobs=all_failed_jobs,
skipped_jobs=all_skipped_jobs)
def get_tool_panel(gi):
tool_client = ToolClient(gi)
return tool_client.get_tool_panel()
parser.read(sys.argv[1])
else:
print "You passed %s I need a .ini file" %(sys.argv[1],)
sys.exit(1)
else:
parser.read('configuration.ini')
api_key = get_api_key(parser.get('Globals', 'api_file'))
galaxy_host = parser.get('Globals', 'galaxy_host')
file_name_re = re.compile(parser.get('Globals', 'sample_re'))
galaxyInstance = GalaxyInstance(galaxy_host, key=api_key)
historyClient = HistoryClient(galaxyInstance)
toolClient = ToolClient(galaxyInstance)
workflowClient = WorkflowClient(galaxyInstance)
dataSetClient = DatasetClient(galaxyInstance)
files = get_files(parser.get('Globals','fastq_dir'))
if len(files) == 0:
print "Not able to find any fastq files looked in %s" %(parser.get('Globals', 'fastq_dir'))
else:
print "Found fastq files running workflow for the following files (R2's will be added)"
print ",".join(files)
files_to_keep = {}
for R1 in files:
input_dir_path = os.path.dirname(R1)+"/"
R2 = R1.replace('R1','R2')
if not os.path.exists(R1):
print "%s File Not Found" % (R1, )
def toolbox(self):
"""
Gets the toolbox elements from <galaxy_url>/api/tools
"""
tool_client = ToolClient(self.gi)
return tool_client.get_tool_panel()
def retrieve_tools_from_instance(cls, instance):
galaxy_instance = GalaxyInstance(url=instance.url)
tool_client = ToolClient(galaxy_instance)
for element in tool_client.get_tools():
if element['model_class'] == 'Tool':
tool_name = element['id']
if '/' in tool_name:
tool_name = tool_name.split('/')[-2]
tool = Tool.query.filter_by(name=tool_name).first()
if tool is None:
tool = Tool(name=tool_name)
db.session.add(tool)
tool.description = element['description']
tool.display_name = element['name']
if 'link' in element:
link = element.get('link', None)
link_start = link.find('/tool_runner')
if link_start != -1:
tool.link = link[link_start:]
for edam_opetation_id in element.get('edam_operations', []):
edam_operation = EDAMOperation.get_from_id(
edam_opetation_id, allow_creation=True)
if edam_operation is not None and edam_operation not in tool.edam_operations:
tool.edam_operations.append(edam_operation)
if 'tool_shed_repository' in element:
tool_version = ToolVersion.query\
.filter_by(name=tool_name)\
.filter_by(changeset=element['tool_shed_repository']['changeset_revision'])\
.filter_by(tool_shed=element['tool_shed_repository']['tool_shed'])\
.filter_by(owner=element['tool_shed_repository']['owner'])\
.first()
else:
tool_version = ToolVersion.query\
.filter_by(name=tool_name)\
.filter_by(version=element['version'])\
.filter_by(tool_shed=None)\
.filter_by(owner=None)\
.first()
if tool_version is None:
tool_version = ToolVersion(name=tool_name,
version=element['version'])
db.session.add(tool_version)
if 'tool_shed_repository' in element:
tool_version.changeset = element['tool_shed_repository'][
'changeset_revision']
tool_version.tool_shed = element['tool_shed_repository'][
'tool_shed']
tool_version.owner = element['tool_shed_repository'][
'owner']
if instance not in tool_version.instances:
tool_version.instances.append(instance)
if tool_version not in tool.versions:
tool.versions.append(tool_version)
db.session.commit()
def retrieve_tools_from_instance(cls, instance):
galaxy_instance = GalaxyInstance(url=instance.url)
tool_client = ToolClient(galaxy_instance)
for element in tool_client.get_tools():
if element['model_class'] == 'Tool':
tool_name = element['id']
if '/' in tool_name:
tool_name = tool_name.split('/')[-2]
try:
tool = Tool.objects.get(name=tool_name)
except Tool.DoesNotExist:
tool = Tool(name=tool_name)
tool.description = element['description']
tool.display_name = element['name']
for edam_opetation_id in element['edam_operations']:
edam_operation = EDAMOperation.get_from_id(
edam_opetation_id, allow_creation=True)
if edam_operation is not None and edam_operation not in tool.edam_operations:
tool.edam_operations.append(edam_operation)
try:
if 'tool_shed_repository' in element:
tool_version = ToolVersion.objects.get(
name=tool_name,
changeset=element['tool_shed_repository']
['changeset_revision'],
tool_shed=element['tool_shed_repository']
['tool_shed'],
owner=element['tool_shed_repository']['owner'])
else:
tool_version = ToolVersion.objects.get(
name=tool_name,
version=element['version'],
tool_shed=None,
owner=None)
except ToolVersion.DoesNotExist:
tool_version = ToolVersion(name=tool_name,
version=element['version'])
if 'tool_shed_repository' in element:
tool_version.changeset = element['tool_shed_repository'][
'changeset_revision']
tool_version.tool_shed = element['tool_shed_repository'][
'tool_shed']
tool_version.owner = element['tool_shed_repository'][
'owner']
if instance not in tool_version.instances:
tool_version.instances.append(instance)
tool_version.save()
if tool_version not in tool.versions:
tool.versions.append(tool_version)
tool.save()
def toolbox(self):
"""
Gets the toolbox elements from <galaxy_url>/api/tools
"""
tool_client = ToolClient(self.gi)
return tool_client.get_tool_panel()
#!/usr/bin/env python
"""
Use the bioblend API to create a fresh history and add a set of files to the history that were imported into the container during the build
Usage: create_and_upload_history.py history_name url1 url2 url3 ...
"""
import sys
from bioblend.galaxy import GalaxyInstance
from bioblend.galaxy.histories import HistoryClient
from bioblend.galaxy.tools import ToolClient
gi = GalaxyInstance(url='http://localhost:80', key='admin')
tc = ToolClient(gi)
lc = HistoryClient(gi)
details = lc.create_history(sys.argv[1])
print "HIST ID: %s" % details["id"]
i = 0
for url in sys.argv:
url_parts = url.split("/")
fname = url_parts[-1]
if i < 2:
i+=1
continue
i+=1
print "submitting %s as %s" % (url,fname)
tc.put_url(url,details["id"],file_name=fname)
def runWorkflow(argDictionary, comparisons):
from bioblend.galaxy import GalaxyInstance
from bioblend.galaxy.histories import HistoryClient
from bioblend.galaxy.tools import ToolClient
from bioblend.galaxy.workflows import WorkflowClient
from bioblend.galaxy.libraries import LibraryClient
import time
api_key = ''
galaxy_host = 'http://localhost:8080/'
gi = GalaxyInstance(url=galaxy_host, key=api_key)
history_client = HistoryClient(gi)
tool_client = ToolClient(gi)
workflow_client = WorkflowClient(gi)
library_client = LibraryClient(gi)
history = history_client.create_history(row['accessionNumber'])
# Import the galaxy workflow
workflow = workflow_client.show_workflow('a799d38679e985db')
input_file = tool_client.upload_file(comparisons, history['id'], file_type='txt')
# Run workflow on csv data to create a new history.
params = dict()
for key in workflow['steps'].keys():
params[key] = argDictionary
datamap = {'1' : {'id': input_file['outputs'][0]['id'], 'src': 'hda'}}
workflow_client.invoke_workflow(workflow['id'], inputs = datamap, history_id = history['id'], params = params)
# A diry hack, we want to wait until we have all datasets
while getNumberNotComplete(history['id'], history_client) > 0:
time.sleep(10)
dataset_id = getFoldChangeData(history, history_client)['id']
return_collection = [{'accessionNo':argDictionary['accessionNumber'], 'foldChange': getUrl(dataset_id),
'PCA': getUrl(getMostRecentDatasetByName('PCAplot.png', history, history_client)['id']),'chrDirTable': getUrl(getMostRecentDatasetByName('chrDirTable.tabular', history, history_client)['id'])}]
number_of_comparisons = -1
for line in open(comparisons):
if not line.isspace():
number_of_comparisons += 1
for comparison in range(0, int(number_of_comparisons)):
tool_inputs = {
'foldChangeTable' : {'id': dataset_id, 'src': 'hda'},
'comparisonNumber' : comparison + 1
}
tool_client.run_tool(history['id'], 'cutFoldChangeTable', tool_inputs)
while getNumberNotComplete(history['id'], history_client) > 0:
time.sleep(10)
if argDictionary['species'] in ["Rat","Cow","Horse","Pig","Zebrafish"]:
pathwayAnalysisWorkflow = workflow_client.show_workflow('c9468fdb6dc5c5f1')
params = dict()
for key in pathwayAnalysisWorkflow['steps'].keys():
params[key] = argDictionary
if argDictionary['species'] == "Rat":
network=getLibraryToolDataID(history=history,history_client=history_client,library_client=library_client,name="ratStringNetwork")
geneLengths=getLibraryToolDataID(history=history,history_client=history_client,library_client=library_client,name="ratGeneLengths")
homology=getLibraryToolDataID(history=history,history_client=history_client,library_client=library_client,name="Homology.rat.txt")
if argDictionary['species'] == "Cow":
network=getLibraryToolDataID(history=history,history_client=history_client,library_client=library_client,name="cowStringNetwork")
geneLengths=getLibraryToolDataID(history=history,history_client=history_client,library_client=library_client,name="cowGeneLengths")
homology=getLibraryToolDataID(history=history,history_client=history_client,library_client=library_client,name="Homology.cow.txt")
if argDictionary['species'] == "Horse":
network=getLibraryToolDataID(history=history,history_client=history_client,library_client=library_client,name="horseStringNetwork")
geneLengths=getLibraryToolDataID(history=history,history_client=history_client,library_client=library_client,name="horseGeneLengths")
homology=getLibraryToolDataID(history=history,history_client=history_client,library_client=library_client,name="Homology.horse.txt")
if argDictionary['species'] == "Pig":
network=getLibraryToolDataID(history=history,history_client=history_client,library_client=library_client,name="pigStringNetwork.txt")
geneLengths=getLibraryToolDataID(history=history,history_client=history_client,library_client=library_client,name="pigGeneLengths.tabular")
homology=getLibraryToolDataID(history=history,history_client=history_client,library_client=library_client,name="Homology.pig.txt")
if argDictionary['species'] == "Zebrafish":
network=getLibraryToolDataID(history=history,history_client=history_client,library_client=library_client,name="zebrafishStringNetwork")
geneLengths=getLibraryToolDataID(history=history,history_client=history_client,library_client=library_client,name="zebrafishGeneLengths")
homology=getLibraryToolDataID(history=history,history_client=history_client,library_client=library_client,name="HOM_AllOrganism.rpt")
pathwayDatamap = {'3' : {'id': homology, 'src': 'hda'},'2' : {'id': network, 'src': 'hda'},'1' : {'id': geneLengths, 'src': 'hda'}}
diffExpDataCollection = getDatasetsByName('cutTable.tabular', history, history_client)
for index, diffExpData in enumerate(diffExpDataCollection):
numCompleted = getNumberComplete(history['id'], history_client) + 10
print(numCompleted)
pathwayDatamap["0"] = {'id': diffExpData['id'], 'src': 'hda'}
workflow_client.invoke_workflow(pathwayAnalysisWorkflow['id'],
inputs = pathwayDatamap,
history_id = history['id'],
params = params)
comparisonDict = getRowFromCsv(comparisons, index)
if 'Factor1' in comparisonDict.keys():
comparisonDict['Factor'] = comparisonDict['Factor1'] + "." + comparisonDict['Factor2']
if 'Paired1' in comparisonDict.keys():
comparisonDict['Factor'] = comparisonDict['Paired1']
return_dict = {'accessionNo':argDictionary['accessionNumber'],
'factor':comparisonDict['Factor'],
'comparisonNum':comparisonDict['Numerator'],
'comparisonDenom':comparisonDict['Denominator'],
'foldChange': getUrl(diffExpData['id']),
'interactome': pathwayDatamap['0']['id'],
'exonLength': pathwayDatamap['2']['id']}
while getNumberComplete(history['id'], history_client) < numCompleted:
time.sleep(10)
return_dict['moduleNodes'] = getUrl(getMostRecentDatasetByName('moduleNodes.text',
history, history_client)['id'])
return_dict['modulePlots'] = getUrl(getMostRecentDatasetByName('modulePlots.pdf',
history, history_client)['id'])
return_dict['slimEnrichmentPathways'] = getUrl(getMostRecentDatasetByName('slimEnrichmentPathways.tabular',
history, history_client)['id'])
return_dict['slimEnrichmentPlot'] = getUrl(getMostRecentDatasetByName('slimEnrichmentPlot.png',
history, history_client)['id'])
return_collection.append(return_dict)
# Hard code keys to define the order
keys = ['accessionNo','factor','comparisonNum','comparisonDenom','PCA','chrDirTable','foldChange',
'interactome','exonLength','moduleNodes','modulePlots','enrichmentTable','slimEnrichmentPathways','slimEnrichmentPlot']
with open('output/' + argDictionary['accessionNumber'] + '-workflowOutput.csv', 'wb') as csvFile:
# Get headers from last dictionary in collection as first doesn't contain all keys
csvOutput = csv.DictWriter(csvFile, keys)
csvOutput.writeheader()
csvOutput.writerows(return_collection)
return return_collection
else:
pathwayAnalysisWorkflow = workflow_client.show_workflow('e85a3be143d5905b')
params = dict()
for key in pathwayAnalysisWorkflow['steps'].keys():
params[key] = argDictionary
# MouseGeneLengths.tab has id 457f69dd7016f307 - step 2 of workflow
# Mouse interactome has id 073be90ac6c3bce5 - step 0 of workflow
if argDictionary['species'] == "Mouse":
network=getLibraryToolDataID(history=history,history_client=history_client,library_client=library_client,name="mouseStringNetwork")
geneLengths=getLibraryToolDataID(history=history,history_client=history_client,library_client=library_client,name="MouseGeneLengths.tab")
homology=getLibraryToolDataID(history=history,history_client=history_client,library_client=library_client,name="Homology.mouse.txt")
secretedReference=getLibraryToolDataID(history=history,history_client=history_client,library_client=library_client,name="uniprot-secreted-mouse.txt")
pathwayDatamap = {'4' : {'id': secretedReference, 'src': 'hda'},'3' : {'id': homology, 'src': 'hda'},'2' : {'id': network, 'src': 'hda'},'1' : {'id': geneLengths, 'src': 'hda'}}
else:
network=getLibraryToolDataID(history=history,history_client=history_client,library_client=library_client,name="humanStringNetwork")
geneLengths=getLibraryToolDataID(history=history,history_client=history_client,library_client=library_client,name="geneLengths")
homology=getLibraryToolDataID(history=history,history_client=history_client,library_client=library_client,name="Homology.mouse.txt")
secretedReference=getLibraryToolDataID(history=history,history_client=history_client,library_client=library_client,name="uniprot-secreted-human.txt")
pathwayDatamap = {'4' : {'id': secretedReference, 'src': 'hda'},'3' : {'id': homology, 'src': 'hda'},'2' : {'id': network, 'src': 'hda'},'1' : {'id': geneLengths, 'src': 'hda'}}
diffExpDataCollection = getDatasetsByName('cutTable.tabular', history, history_client)
for index, diffExpData in enumerate(diffExpDataCollection):
numCompleted = getNumberComplete(history['id'], history_client) + 14
print(numCompleted)
pathwayDatamap["0"] = {'id': diffExpData['id'], 'src': 'hda'}
workflow_client.invoke_workflow(pathwayAnalysisWorkflow['id'],
inputs = pathwayDatamap,
history_id = history['id'],
params = params)
comparisonDict = getRowFromCsv(comparisons, index)
if 'Factor1' in comparisonDict.keys():
comparisonDict['Factor'] = comparisonDict['Factor1'] + "." + comparisonDict['Factor2']
if 'Paired1' in comparisonDict.keys():
comparisonDict['Factor'] = comparisonDict['Paired1']
return_dict = {'accessionNo':argDictionary['accessionNumber'],
'factor':comparisonDict['Factor'],
'comparisonNum':comparisonDict['Numerator'],
'comparisonDenom':comparisonDict['Denominator'],
'foldChange': getUrl(diffExpData['id']),
'interactome': pathwayDatamap['0']['id'],
'exonLength': pathwayDatamap['2']['id']}
while getNumberComplete(history['id'], history_client) < numCompleted:
time.sleep(10)
return_dict['moduleNodes'] = getUrl(getMostRecentDatasetByName('moduleNodes.text',
history, history_client)['id'])
return_dict['modulePlots'] = getUrl(getMostRecentDatasetByName('modulePlots.pdf',
history, history_client)['id'])
return_dict['pathways'] = getUrl(getMostRecentDatasetByName('pathways.tabular',
history, history_client)['id'])
return_dict['enrichPlot'] = getUrl(getMostRecentDatasetByName('enrichmentPlot.png',
history, history_client)['id'])
return_dict['enrichmentTable'] = getUrl(getMostRecentDatasetByName('TF_EnrichmentTable.tabular',
history, history_client)['id'])
return_dict['slimEnrichmentPathways'] = getUrl(getMostRecentDatasetByName('slimEnrichmentPathways.tabular',
history, history_client)['id'])
return_dict['slimEnrichmentPlot'] = getUrl(getMostRecentDatasetByName('slimEnrichmentPlot.png',
history, history_client)['id'])
return_collection.append(return_dict)
# Hard code keys to define the order
keys = ['accessionNo','factor','comparisonNum','comparisonDenom','PCA','chrDirTable','foldChange',
'interactome','exonLength','moduleNodes','modulePlots','pathways','enrichPlot','enrichmentTable','slimEnrichmentPathways','slimEnrichmentPlot']
with open('output/' + argDictionary['accessionNumber'] + '-workflowOutput.csv', 'wb') as csvFile:
# Get headers from last dictionary in collection as first doesn't contain all keys
csvOutput = csv.DictWriter(csvFile, keys)
csvOutput.writeheader()
csvOutput.writerows(return_collection)
return return_collection
def get_tool_panel(gi):
tool_client = ToolClient(gi)
return tool_client.get_tool_panel()
if __name__ == '__main__':
# GET PATH NAMES AND EXTENSIONS FROM COMMAND LINE INPUT
input_file_full = sys.argv[1]
input_file_format = input_file_full[input_file_full.rfind(".")+1:len(input_file_full)]
output_file_full = sys.argv[2]
output_file_format = output_file_full[output_file_full.rfind(".")+1:len(output_file_full)]
# CHOOSE CONVERTER
tool_id = choose_converter(input_file_format,output_file_format)
# INITIALIZE GALAXY
galaxy_instance = GalaxyInstance(url=base_url, key=apikey)
history_client = HistoryClient(galaxy_instance)
tool_client = ToolClient(galaxy_instance)
dataset_client = DatasetClient(galaxy_instance)
history = history_client.create_history('tmp')
# UPLOAD FILES
input_file_1 = tool_client.upload_file(input_file_full, history['id'], type='txt')
input_file_2 = tool_client.upload_file(input_file_full, history['id'], type='txt')
params = {'input_numbers_001':{'src': 'hda', 'id': input_file_1['outputs'][0]['id']},'input_numbers_002':{'src': 'hda', 'id': input_file_2['outputs'][0]['id']}}
wait_4_process(history['id'],"uploading files")
# RUN CONVERSION
runtool_output = tool_client.run_tool(history_id=history['id'], tool_id=tool_id, tool_inputs=params)
wait_4_process(history['id'],"running tool")
# DOWNLOAD CONVERTED FILE
download_output = dataset_client.download_dataset(runtool_output['jobs'][0]['id'],output_file_full, use_default_filename=False)
def get_tools(gi):
tool_client = ToolClient(gi)
return tool_client.get_tools()
