def map_experiment_files(project_id, datasetPath, mapping):
files = builder_utils.listDirectoryFiles(datasetPath)
for file in files:
outputfile = os.path.join(datasetPath, file)
data = builder_utils.readDataset(outputfile)
data = map_experimental_data(data, mapping)
builder_utils.export_contents(data, datasetPath, file)
def create_experiment_internal_identifiers(driver, projectId, data, directory, filename):
done = 0
df = create_new_subjects(driver, data, projectId)
df1 = create_new_biosamples(driver, df)
df2 = create_new_ansamples(driver, df1)
builder_utils.export_contents(df2, directory, filename)
done += 1
return done
def create_mapping_cols_clinical(driver,
data,
directory,
filename,
separator='|'):
"""
:param driver: neo4j driver, which provides the connection to the neo4j graph database.
:type driver: neo4j driver
:param data: pandas Dataframe with clinical data as columns and samples as rows.
:param str separator: character used to separate multiple entries in an attribute.
:return: Pandas Dataframe with all clinical data and graph database internal identifiers.
"""
tissue_dict = {}
disease_dict = {}
intervention_dict = {}
if 'disease' in data:
for disease in data['disease'].dropna().unique():
if len(disease.split(separator)) > 1:
ids = []
for i in disease.split(separator):
disease_id = query_utils.map_node_name_to_id(
driver, 'Disease', str(i.strip()))
if disease_id is not None:
ids.append(disease_id)
disease_dict[disease] = '|'.join(ids)
else:
disease_id = query_utils.map_node_name_to_id(
driver, 'Disease', str(disease.strip()))
disease_dict[disease] = disease_id
data['disease id'] = data['disease'].map(disease_dict)
if 'tissue' in data:
for tissue in data['tissue'].dropna().unique():
tissue_id = query_utils.map_node_name_to_id(
driver, 'Tissue', str(tissue.strip()))
tissue_dict[tissue] = tissue_id
data['tissue id'] = data['tissue'].map(tissue_dict)
if 'studies_intervention' in data:
for interventions in data['studies_intervention'].dropna().unique():
for intervention in str(interventions).split('|'):
if len(intervention.split()) > 1:
intervention_dict[intervention] = re.search(
r'\(([^)]+)',
intervention.split()[-1]).group(1)
else:
intervention_dict[intervention] = intervention
data['intervention id'] = data['studies_intervention'].map(
intervention_dict)
builder_utils.export_contents(data, directory, filename)
def save_files_in_tmp(content, dataset, prot_tool, prot_file, projectid,
uploaded_file):
if dataset is not None:
session_cookie = flask.request.cookies.get('custom-auth-session')
temporaryDirectory = os.path.join(tmpDirectory,
session_cookie + "upload")
if not os.path.exists(tmpDirectory):
os.makedirs(tmpDirectory)
elif not os.path.exists(temporaryDirectory):
os.makedirs(temporaryDirectory)
directory = os.path.join(temporaryDirectory, dataset)
if os.path.exists(directory) and uploaded_file is not None:
if os.path.exists(os.path.join(directory, uploaded_file)):
shutil.rmtree(directory)
builder_utils.checkDirectory(directory)
if dataset in ['proteomics', 'interactomics', 'phosphoproteomics'
] and prot_tool != '' and prot_file != '':
selected_file = prot_tool.lower() + "-" + prot_file.lower()
if selected_file in config['file_proteomics']:
filename = config['file_proteomics'][selected_file]
else:
filename = dataset + '_' + prot_tool.lower(
) + '_' + prot_file.replace(
' ', '').lower() + '.' + uploaded_file.split('.')[-1]
directory = os.path.join(directory, prot_tool.lower())
if os.path.exists(directory):
if os.path.exists(os.path.join(directory, filename)):
os.remove(os.path.join(directory, filename))
builder_utils.checkDirectory(directory)
elif dataset == 'experimental_design':
filename = config['file_design'].split(
'_')[0] + '_' + projectid + '.' + uploaded_file.split('.')[-1]
elif dataset == 'clinical':
filename = config['file_clinical'].split(
'_')[0] + '_' + projectid + '.' + uploaded_file.split('.')[-1]
if uploaded_file is None:
content = None
if content is not None:
data = builder_utils.parse_contents(content, filename)
builder_utils.export_contents(data, directory, filename)
uploaded = uploaded_file
uploaded_file = None
return uploaded, uploaded_file, '', ''
else:
raise PreventUpdate
return '', None, '', ''