def insertDrugs(uri, keyspace, num_dr, num_threads, ctn, session):
file = '../biograkn-covid/Dataset/DGIdb/dgidb_drugs.tsv'
print(' ')
print('Opening DGIdb...')
print(' ')
raw_file = openFile(file, num_dr)
drugs = []
for i in raw_file[:num_dr]:
data = {}
data['drug-claim-name'] = i[0].strip('"')
data['drug-name'] = i[1].strip('"')
data['chembl-id'] = i[2]
data['drug-claim-source'] = i[3]
drugs.append(data)
counter = 0
drugs_list = drugs
batches = []
batches2 = []
tx = session.transaction().write()
pool = ThreadPool(num_threads)
for d in drugs_list:
counter = counter + 1
graql = f'''insert $d isa drug, has drug-claim-name "{d['drug-claim-name']}", has drug-name "{d['drug-name']}", has chembl-id "{d['chembl-id']}", has drug-claim-source "{d['drug-claim-source']}";'''
batches.append(graql)
del graql
if counter % ctn == 0:
batches2.append(batches)
batches = []
batches2.append(batches)
pool.map(partial(batch_job, session), batches2)
pool.close()
pool.join()
print('Drugs committed!')
def filterHomoSapiens(num_path):
file = 'Dataset/Reactome/UniProt2Reactome_All_Levels.tsv'
print(' ')
print('Opening Reactome...')
print(' ')
raw_file = openFile(file, num_path)
pathway_associations = []
for i in raw_file[:num_path]:
if i[5] == "H**o sapiens":
data = {}
data['uniprot-id'] = i[0].strip('"')
data['pathway-id'] = i[1].strip('"')
data['pathway-name'] = i[3]
data['organism'] = i[5]
pathway_associations.append(data)
return pathway_associations
def insertInteractions(uri, database, num_int, num_threads, ctn, session):
batches_pr = []
ssl._create_default_https_context = ssl._create_unverified_context
url = "https://www.dgidb.org/data/monthly_tsvs/2021-Jan/interactions.tsv"
wget.download(url, 'Dataset/DGIdb/')
file = 'Dataset/DGIdb/interactions.tsv'
print(' ')
print('Opening DGIdb-Interactions...')
print(' ')
raw_file = openFile(file, num_int)
interactions = []
for i in raw_file[:num_int]:
data = {}
data['gene-name'] = i[0]
data['entrez-id'] = i[2]
data['interaction-type'] = i[4]
data['drug-claim-name'] = i[5]
data['drug-name'] = i[7]
data['chembl-id'] = i[8]
interactions.append(data)
os.remove('Dataset/DGIdb/interactions.tsv')
counter = 0
pool = ThreadPool(num_threads)
batches = []
for q in interactions:
if q['entrez-id'] is not "":
counter = counter + 1
graql = f"""match $g isa gene, has entrez-id "{q['entrez-id']}"; $d isa drug, has drug-claim-name "{q['drug-claim-name']}";"""
# TODO Insert interaction type as a role
if q['interaction-type'] == "":
graql = graql + f"insert $r (target-gene: $g, interacting-drug: $d) isa drug-gene-interaction;"
else:
graql = graql + f"""insert $r (target-gene: $g, interacting-drug: $d) isa drug-gene-interaction, has interaction-type "{q['interaction-type']}";"""
batches.append(graql)
del graql
if counter % ctn == 0:
batches_pr.append(batches)
batches = []
batches_pr.append(batches)
pool.map(partial(batch_job, session), batches_pr)
pool.close()
pool.join()
print('.....')
print('Finished migrating Drug Interactions.')
print('.....')
def filterHomoSapiens(num_path):
url = "https://reactome.org/download/current/UniProt2Reactome_All_Levels.txt"
wget.download(url, 'Dataset/Reactome/')
file = 'Dataset/Reactome/UniProt2Reactome_All_Levels.txt'
print(' ')
print('Opening Reactome...')
print(' ')
raw_file = openFile(file, num_path)
pathway_associations = []
for i in raw_file[:num_path]:
if i[5] == "H**o sapiens":
data = {}
data['uniprot-id'] = i[0].strip('"')
data['pathway-id'] = i[1].strip('"')
data['pathway-name'] = i[3]
data['organism'] = i[5]
pathway_associations.append(data)
return pathway_associations
def insertInteractions(uri, keyspace, num_int, num_threads, ctn, session):
batches_pr = []
file = '../biograkn-covid/Dataset/DGIdb/dgidb_interactions.tsv'
print(' ')
print('Opening DGIdb-Interactions...')
print(' ')
raw_file = openFile(file, num_int)
interactions = []
for i in raw_file[:num_int]:
data = {}
data['gene-name'] = i[0]
data['entrez-id'] = i[2]
data['interaction-type'] = i[4]
data['drug-claim-name'] = i[5]
data['drug-name'] = i[7]
data['chembl-id'] = i[8]
interactions.append(data)
counter = 0
pool = ThreadPool(num_threads)
batches = []
for q in interactions:
if q['entrez-id'] is not "":
counter = counter + 1
graql = f"""match $g isa gene, has entrez-id "{q['entrez-id']}"; $d isa drug, has drug-claim-name "{q['drug-claim-name']}";"""
# TODO Insert interaction type as a role
if q['interaction-type'] == "":
graql = graql + f"insert $r (target-gene: $g, interacting-drug: $d) isa drug-gene-interaction;"
else:
graql = graql + f"""insert $r (target-gene: $g, interacting-drug: $d) isa drug-gene-interaction, has interaction-type "{q['interaction-type']}";"""
batches.append(graql)
del graql
if counter % ctn == 0:
batches_pr.append(batches)
batches = []
batches_pr.append(batches)
pool.map(partial(batch_job, session), batches_pr)
pool.close()
pool.join()
print('.....')
print('Finished migrating Drug Interactions.')
print('.....')
def insert_interactions(session, num_int, num_threads, batch_size):
print(' Downloading drug-gene interactions dataset')
ssl._create_default_https_context = ssl._create_unverified_context
url = "https://www.dgidb.org/data/monthly_tsvs/2021-Jan/interactions.tsv"
wget.download(url, 'Dataset/DGIdb/')
print(' Finished downloading')
file = 'Dataset/DGIdb/interactions.tsv'
raw_file = openFile(file, num_int)
interactions = []
for i in raw_file[:num_int]:
data = {}
data['gene-name'] = i[0]
data['entrez-id'] = i[2]
data['interaction-type'] = i[4]
data['drug-claim-name'] = i[5]
data['drug-name'] = i[7]
data['chembl-id'] = i[8]
interactions.append(data)
os.remove('Dataset/DGIdb/interactions.tsv')
print(' Starting with drug-gene interactions.')
batches = []
batch = []
total = 0
for q in interactions:
if q['entrez-id'] != "":
typeql = f"""match $g isa gene, has entrez-id "{q['entrez-id']}"; $d isa drug, has drug-claim-name "{q['drug-claim-name']}";"""
# TODO Insert interaction type as a role
if q['interaction-type'] == "":
typeql = typeql + f"insert $r (target-gene: $g, interacting-drug: $d) isa drug-gene-interaction;"
else:
typeql = typeql + f"""insert $r (target-gene: $g, interacting-drug: $d) isa drug-gene-interaction, has interaction-type "{q['interaction-type']}";"""
batch.append(typeql)
total += 1
if len(batch) == batch_size:
batches.append(batch)
batch = []
batches.append(batch)
pool = ThreadPool(num_threads)
pool.map(partial(write_batch, session), batches)
pool.close()
pool.join()
print(f' Finished drug-gene interactions. ({total} entries) ')
def insertDrugs(uri, database, num_dr, num_threads, ctn, session):
#from Migrators.Helpers.get_file import get_file
get_file("https://www.dgidb.org/data/monthly_tsvs/2021-Jan/drugs.tsv",
"Dataset/DGIdb/")
file = 'Dataset/DGIdb/drugs.tsv'
print(' ')
print('Opening DGIdb...')
print(' ')
raw_file = openFile(file, num_dr)
drugs = []
for i in raw_file[:num_dr]:
data = {}
data['drug-claim-name'] = i[0].strip('"')
data['drug-name'] = i[1].strip('"')
data['chembl-id'] = i[2]
data['drug-claim-source'] = i[3]
drugs.append(data)
os.remove('Dataset/DGIdb/drugs.tsv')
counter = 0
drugs_list = drugs
batches = []
batches2 = []
pool = ThreadPool(num_threads)
for d in drugs_list:
counter = counter + 1
graql = f'''insert $d isa drug, has drug-claim-name "{d['drug-claim-name']}", has drug-name "{d['drug-name']}", has chembl-id "{d['chembl-id']}", has drug-claim-source "{d['drug-claim-source']}";'''
batches.append(graql)
del graql
if counter % ctn == 0:
batches2.append(batches)
batches = []
batches2.append(batches)
pool.map(partial(batch_job, session), batches2)
pool.close()
pool.join()
print('Drugs committed!')
def insert_drugs(session, num_dr, num_threads, batch_size):
# from Migrators.Helpers.get_file import get_file
print(' Downloading dataset')
get_file("https://www.dgidb.org/data/monthly_tsvs/2021-Jan/drugs.tsv",
"Dataset/DGIdb/")
print(' Finished downloading')
file = 'Dataset/DGIdb/drugs.tsv'
raw_file = openFile(file, num_dr)
drugs = []
for i in raw_file[:num_dr]:
data = {}
data['drug-claim-name'] = i[0].strip('"')
data['drug-name'] = i[1].strip('"')
data['chembl-id'] = i[2]
data['drug-claim-source'] = i[3]
drugs.append(data)
os.remove('Dataset/DGIdb/drugs.tsv')
drugs_list = drugs
print(' Starting with drugs.')
batch = []
batches = []
total = 0
for d in drugs_list:
typeql = f'''insert $d isa drug, has drug-claim-name "{d['drug-claim-name']}", has drug-name "{d['drug-name']}", has chembl-id "{d['chembl-id']}", has drug-claim-source "{d['drug-claim-source']}";'''
batch.append(typeql)
total += 1
if len(batch) == batch_size:
batches.append(batch)
batch = []
batches.append(batch)
pool = ThreadPool(num_threads)
pool.map(partial(write_batch, session), batches)
pool.close()
pool.join()
print(f' Drugs inserted! ({total} entries)')
