From db3da4acafc6a775b51decced904b5ab0edc4724 Mon Sep 17 00:00:00 2001
From: AndiMajore <andi.majore@googlemail.com>
Date: Thu, 30 Jun 2022 17:49:50 +0200
Subject: [PATCH] nedrex integration and cleanup progressed
Former-commit-id: e064c2ed7ee312a0e6efb5ce07125e7cd2c38fa7 [formerly ee0efdbd94033f23904dae682523e5f0adf5fac7]
Former-commit-id: 2f82b1ea117749164c5a9241805dfe160cf1effb
---
Dockerfile | 2 +-
.../Disorders/disorders.tsv | 0
.../drugbank-drug_disorder_indication.tsv | 0
{data_drugstone => data}/Drugs/drug-file.txt | 0
.../Drugs/drugbank_labels.csv | 0
.../Expression/gene_tissue_expression.gct | 0
.../PDI/BIOGRID-CHEMICALS-3.5.187.chemtab.txt | 0
.../PDI/DGIdb_drug_gene_interactions.csv | 0
.../PDI/chembl_drug_gene_interactions.csv | 0
.../chembl_drug_gene_interactions_uniq.csv | 0
.../PDI/drugbank_drug_gene_interactions.csv | 0
.../drugbank_drug_gene_interactions_uniq.csv | 0
.../disgenet-protein_disorder_association.tsv | 0
.../PPDr-temp.graphml.REMOVED.git-id | 0
.../PPI-temp.graphml.REMOVED.git-id | 0
...o_sapiens-3.5.187.mitab.txt.REMOVED.git-id | 0
{data_drugstone => data}/PPI/apid_9606_Q2.txt | 0
...tome_homo-sapiens-protein-interactions.txt | 0
.../PPI/string_interactions.csv | 0
.../Proteins/entrez_to_ensg.json | 0
.../Proteins/protein_list.csv | 0
.../drug-protein-interaction.txt | 0
{data_drugstone => data}/protein-file.txt | 0
.../protein_protein_interaction_file.txt | 0
.../temp-PPDr.graphml.REMOVED.git-id | 0
docker-compose.yml | 23 +-
drugstone/backend_tasks.py | 2 +-
.../management/commands/import_from_nedrex.py | 201 +++++++---
drugstone/management/commands/make_graphs.py | 4 +-
drugstone/management/commands/populate_db.py | 176 +++++----
drugstone/management/commands/test.py | 20 +-
drugstone/management/includes/DataLoader.py | 20 +-
.../management/includes/DataPopulator.py | 356 +++++++-----------
.../management/includes/DatasetLoader.py | 99 +++++
drugstone/management/includes/NodeCache.py | 60 +++
drugstone/models.py | 82 ++--
.../docker-entrypoint.sh | 10 +-
scripts/import-data.sh | 16 +-
tasks/betweenness_centrality_test.py | 2 +-
tasks/closeness_centrality_test.py | 2 +-
tasks/degree_centrality_test.py | 2 +-
tasks/keypathwayminer_test.py | 2 +-
tasks/multi_steiner_test.py | 2 +-
tasks/network_proximity_test.py | 2 +-
tasks/task_hook.py | 2 +-
tasks/task_test.py | 2 +-
tasks/trust_rank_test.py | 2 +-
47 files changed, 681 insertions(+), 408 deletions(-)
rename {data_drugstone => data}/Disorders/disorders.tsv (100%)
rename {data_drugstone => data}/DrDi/drugbank-drug_disorder_indication.tsv (100%)
rename {data_drugstone => data}/Drugs/drug-file.txt (100%)
rename {data_drugstone => data}/Drugs/drugbank_labels.csv (100%)
rename {data_drugstone => data}/Expression/gene_tissue_expression.gct (100%)
rename {data_drugstone => data}/PDI/BIOGRID-CHEMICALS-3.5.187.chemtab.txt (100%)
rename {data_drugstone => data}/PDI/DGIdb_drug_gene_interactions.csv (100%)
rename {data_drugstone => data}/PDI/chembl_drug_gene_interactions.csv (100%)
rename {data_drugstone => data}/PDI/chembl_drug_gene_interactions_uniq.csv (100%)
rename {data_drugstone => data}/PDI/drugbank_drug_gene_interactions.csv (100%)
rename {data_drugstone => data}/PDI/drugbank_drug_gene_interactions_uniq.csv (100%)
rename {data_drugstone => data}/PDi/disgenet-protein_disorder_association.tsv (100%)
rename {data_drugstone => data}/PPDr-temp.graphml.REMOVED.git-id (100%)
rename {data_drugstone => data}/PPI-temp.graphml.REMOVED.git-id (100%)
rename {data_drugstone => data}/PPI/BIOGRID-ORGANISM-Homo_sapiens-3.5.187.mitab.txt.REMOVED.git-id (100%)
rename {data_drugstone => data}/PPI/apid_9606_Q2.txt (100%)
rename {data_drugstone => data}/PPI/reactome_homo-sapiens-protein-interactions.txt (100%)
rename {data_drugstone => data}/PPI/string_interactions.csv (100%)
rename {data_drugstone => data}/Proteins/entrez_to_ensg.json (100%)
rename {data_drugstone => data}/Proteins/protein_list.csv (100%)
rename {data_drugstone => data}/drug-protein-interaction.txt (100%)
rename {data_drugstone => data}/protein-file.txt (100%)
rename {data_drugstone => data}/protein_protein_interaction_file.txt (100%)
rename {data_drugstone => data}/temp-PPDr.graphml.REMOVED.git-id (100%)
create mode 100644 drugstone/management/includes/DatasetLoader.py
create mode 100644 drugstone/management/includes/NodeCache.py
rename docker-entrypoint.sh => scripts/docker-entrypoint.sh (63%)
diff --git a/Dockerfile b/Dockerfile
index aec4578..5bafbf3 100755
--- a/Dockerfile
+++ b/Dockerfile
@@ -22,7 +22,7 @@ RUN pip install -r /usr/src/drugstone/requirements.txt
RUN pip install gunicorn
COPY ./supervisord.conf /etc/supervisor/conf.d/supervisord.conf
-COPY ./docker-entrypoint.sh /usr/src/drugstone/docker-entrypoint.sh
+#COPY scripts/docker-entrypoint.sh /usr/src/drugstone/docker-entrypoint.sh
# COPY ./scripts/ /usr/src/drugstone/scripts/
COPY ./python_nedrex/ /usr/src/drugstone/python_nedrex/
RUN pip install /usr/src/drugstone/python_nedrex/
diff --git a/data_drugstone/Disorders/disorders.tsv b/data/Disorders/disorders.tsv
similarity index 100%
rename from data_drugstone/Disorders/disorders.tsv
rename to data/Disorders/disorders.tsv
diff --git a/data_drugstone/DrDi/drugbank-drug_disorder_indication.tsv b/data/DrDi/drugbank-drug_disorder_indication.tsv
similarity index 100%
rename from data_drugstone/DrDi/drugbank-drug_disorder_indication.tsv
rename to data/DrDi/drugbank-drug_disorder_indication.tsv
diff --git a/data_drugstone/Drugs/drug-file.txt b/data/Drugs/drug-file.txt
similarity index 100%
rename from data_drugstone/Drugs/drug-file.txt
rename to data/Drugs/drug-file.txt
diff --git a/data_drugstone/Drugs/drugbank_labels.csv b/data/Drugs/drugbank_labels.csv
similarity index 100%
rename from data_drugstone/Drugs/drugbank_labels.csv
rename to data/Drugs/drugbank_labels.csv
diff --git a/data_drugstone/Expression/gene_tissue_expression.gct b/data/Expression/gene_tissue_expression.gct
similarity index 100%
rename from data_drugstone/Expression/gene_tissue_expression.gct
rename to data/Expression/gene_tissue_expression.gct
diff --git a/data_drugstone/PDI/BIOGRID-CHEMICALS-3.5.187.chemtab.txt b/data/PDI/BIOGRID-CHEMICALS-3.5.187.chemtab.txt
similarity index 100%
rename from data_drugstone/PDI/BIOGRID-CHEMICALS-3.5.187.chemtab.txt
rename to data/PDI/BIOGRID-CHEMICALS-3.5.187.chemtab.txt
diff --git a/data_drugstone/PDI/DGIdb_drug_gene_interactions.csv b/data/PDI/DGIdb_drug_gene_interactions.csv
similarity index 100%
rename from data_drugstone/PDI/DGIdb_drug_gene_interactions.csv
rename to data/PDI/DGIdb_drug_gene_interactions.csv
diff --git a/data_drugstone/PDI/chembl_drug_gene_interactions.csv b/data/PDI/chembl_drug_gene_interactions.csv
similarity index 100%
rename from data_drugstone/PDI/chembl_drug_gene_interactions.csv
rename to data/PDI/chembl_drug_gene_interactions.csv
diff --git a/data_drugstone/PDI/chembl_drug_gene_interactions_uniq.csv b/data/PDI/chembl_drug_gene_interactions_uniq.csv
similarity index 100%
rename from data_drugstone/PDI/chembl_drug_gene_interactions_uniq.csv
rename to data/PDI/chembl_drug_gene_interactions_uniq.csv
diff --git a/data_drugstone/PDI/drugbank_drug_gene_interactions.csv b/data/PDI/drugbank_drug_gene_interactions.csv
similarity index 100%
rename from data_drugstone/PDI/drugbank_drug_gene_interactions.csv
rename to data/PDI/drugbank_drug_gene_interactions.csv
diff --git a/data_drugstone/PDI/drugbank_drug_gene_interactions_uniq.csv b/data/PDI/drugbank_drug_gene_interactions_uniq.csv
similarity index 100%
rename from data_drugstone/PDI/drugbank_drug_gene_interactions_uniq.csv
rename to data/PDI/drugbank_drug_gene_interactions_uniq.csv
diff --git a/data_drugstone/PDi/disgenet-protein_disorder_association.tsv b/data/PDi/disgenet-protein_disorder_association.tsv
similarity index 100%
rename from data_drugstone/PDi/disgenet-protein_disorder_association.tsv
rename to data/PDi/disgenet-protein_disorder_association.tsv
diff --git a/data_drugstone/PPDr-temp.graphml.REMOVED.git-id b/data/PPDr-temp.graphml.REMOVED.git-id
similarity index 100%
rename from data_drugstone/PPDr-temp.graphml.REMOVED.git-id
rename to data/PPDr-temp.graphml.REMOVED.git-id
diff --git a/data_drugstone/PPI-temp.graphml.REMOVED.git-id b/data/PPI-temp.graphml.REMOVED.git-id
similarity index 100%
rename from data_drugstone/PPI-temp.graphml.REMOVED.git-id
rename to data/PPI-temp.graphml.REMOVED.git-id
diff --git a/data_drugstone/PPI/BIOGRID-ORGANISM-Homo_sapiens-3.5.187.mitab.txt.REMOVED.git-id b/data/PPI/BIOGRID-ORGANISM-Homo_sapiens-3.5.187.mitab.txt.REMOVED.git-id
similarity index 100%
rename from data_drugstone/PPI/BIOGRID-ORGANISM-Homo_sapiens-3.5.187.mitab.txt.REMOVED.git-id
rename to data/PPI/BIOGRID-ORGANISM-Homo_sapiens-3.5.187.mitab.txt.REMOVED.git-id
diff --git a/data_drugstone/PPI/apid_9606_Q2.txt b/data/PPI/apid_9606_Q2.txt
similarity index 100%
rename from data_drugstone/PPI/apid_9606_Q2.txt
rename to data/PPI/apid_9606_Q2.txt
diff --git a/data_drugstone/PPI/reactome_homo-sapiens-protein-interactions.txt b/data/PPI/reactome_homo-sapiens-protein-interactions.txt
similarity index 100%
rename from data_drugstone/PPI/reactome_homo-sapiens-protein-interactions.txt
rename to data/PPI/reactome_homo-sapiens-protein-interactions.txt
diff --git a/data_drugstone/PPI/string_interactions.csv b/data/PPI/string_interactions.csv
similarity index 100%
rename from data_drugstone/PPI/string_interactions.csv
rename to data/PPI/string_interactions.csv
diff --git a/data_drugstone/Proteins/entrez_to_ensg.json b/data/Proteins/entrez_to_ensg.json
similarity index 100%
rename from data_drugstone/Proteins/entrez_to_ensg.json
rename to data/Proteins/entrez_to_ensg.json
diff --git a/data_drugstone/Proteins/protein_list.csv b/data/Proteins/protein_list.csv
similarity index 100%
rename from data_drugstone/Proteins/protein_list.csv
rename to data/Proteins/protein_list.csv
diff --git a/data_drugstone/drug-protein-interaction.txt b/data/drug-protein-interaction.txt
similarity index 100%
rename from data_drugstone/drug-protein-interaction.txt
rename to data/drug-protein-interaction.txt
diff --git a/data_drugstone/protein-file.txt b/data/protein-file.txt
similarity index 100%
rename from data_drugstone/protein-file.txt
rename to data/protein-file.txt
diff --git a/data_drugstone/protein_protein_interaction_file.txt b/data/protein_protein_interaction_file.txt
similarity index 100%
rename from data_drugstone/protein_protein_interaction_file.txt
rename to data/protein_protein_interaction_file.txt
diff --git a/data_drugstone/temp-PPDr.graphml.REMOVED.git-id b/data/temp-PPDr.graphml.REMOVED.git-id
similarity index 100%
rename from data_drugstone/temp-PPDr.graphml.REMOVED.git-id
rename to data/temp-PPDr.graphml.REMOVED.git-id
diff --git a/docker-compose.yml b/docker-compose.yml
index 310e7e9..0ea123a 100755
--- a/docker-compose.yml
+++ b/docker-compose.yml
@@ -6,13 +6,14 @@ services:
container_name: drugstone_backend
command:
- "sh"
- - "/usr/src/drugstone/docker-entrypoint.sh"
+ - "scripts/docker-entrypoint.sh"
build: .
env_file:
- 'docker-django.env.dev'
restart: always
volumes:
- - drugstone_backend_volume:/usr/src/drugstone/drugstone/migrations
+ - drugstone_db_schema_volume:/usr/src/drugstone/drugstone/migrations
+ - drugstone_data_volume:/usr/src/drugstone/data
ports:
- 8001:8000
networks:
@@ -30,7 +31,7 @@ services:
networks:
- drugstone_net
volumes:
- - drugstone_postgres_volume:/var/lib/postgresql/data_drugstone/
+ - drugstone_db_volume:/var/lib/postgresql/data
environment:
- POSTGRES_DB=drugstone
- POSTGRES_USER=drugstone
@@ -53,15 +54,13 @@ services:
celery:
command:
- "sh"
- - "/usr/src/drugstone/scripts/start_celery_worker.sh"
+ - "scripts/start_celery_worker.sh"
restart: always
image: drugstone_backend
container_name: drugstone_celery
hostname: drugstone_celery
env_file:
- './docker-django.env.dev'
-# volumes:
-# - ./:/usr/src/drugstone/
depends_on:
- redis
- db
@@ -70,14 +69,12 @@ services:
celery-beat:
command:
- "sh"
- - "/usr/src/drugstone/scripts/start_celery_beat.sh"
+ - "scripts/start_celery_beat.sh"
image: drugstone_backend
container_name: drugstone_celery_beat
hostname: drugstone_celery_beat
env_file:
- './docker-django.env.dev'
-# volumes:
-# - ./:/usr/src/drugstone/
depends_on:
- redis
- db
@@ -97,5 +94,9 @@ networks:
drugstone_net:
volumes:
- drugstone_postgres_volume:
- drugstone_backend_volume:
\ No newline at end of file
+ drugstone_db_volume:
+ external: true
+ drugstone_db_schema_volume:
+ external: true
+ drugstone_data_volume:
+ external: true
\ No newline at end of file
diff --git a/drugstone/backend_tasks.py b/drugstone/backend_tasks.py
index e8a6134..01f7827 100755
--- a/drugstone/backend_tasks.py
+++ b/drugstone/backend_tasks.py
@@ -40,7 +40,7 @@ def run_task(token, algorithm, parameters):
r.set(f'{token}_job_id', f'{job_id}')
r.set(f'{token}_started_at', str(datetime.now().timestamp()))
- task_hook = TaskHook(json.loads(parameters), './data_drugstone/Networks/', set_progress, set_result)
+ task_hook = TaskHook(json.loads(parameters), './data/Networks/', set_progress, set_result)
try:
if algorithm == 'dummy':
diff --git a/drugstone/management/commands/import_from_nedrex.py b/drugstone/management/commands/import_from_nedrex.py
index dc3f052..b3af226 100644
--- a/drugstone/management/commands/import_from_nedrex.py
+++ b/drugstone/management/commands/import_from_nedrex.py
@@ -4,6 +4,7 @@ import python_nedrex as nedrex
from python_nedrex.core import get_nodes, get_edges, get_api_key
from drugstone import models
+from drugstone.management.includes.NodeCache import NodeCache
def iter_node_collection(coll_name, eval):
@@ -35,7 +36,7 @@ def identify_updates(new_list, old_list):
c = list()
for id in new_list:
if id not in old_list:
- c.append(id)
+ c.append(new_list[id])
elif new_list[id] != old_list[id]:
old_list[id].update(new_list[id])
u.append(old_list[id])
@@ -45,51 +46,33 @@ def identify_updates(new_list, old_list):
def format_list(l):
if l is not None and len(l) > 0:
s = str(l)[1:]
- return s[:len(s) - 1].replace("'","")
+ return s[:len(s) - 1].replace("'", "")
return ""
-class nedrex_importer:
- proteins = dict()
- entrez_to_uniprot = dict()
- gene_name_to_uniprot = defaultdict(lambda: set())
- disorders = dict()
- drugs = dict()
+def to_id(string):
+ idx = string.index('.')
+ return string[idx + 1:]
- def __init__(self, base_url):
+
+class NedrexImporter:
+ cache: NodeCache = None
+
+ def __init__(self, base_url, cache: NodeCache):
+ self.cache = cache
nedrex.config.set_url_base(base_url)
api_key = get_api_key(accept_eula=True)
nedrex.config.set_api_key(api_key)
- def init_proteins(self):
- if len(self.proteins) == 0:
- print("Generating protein maps...")
- for protein in models.Protein.objects.all():
- self.proteins[protein.entrez] = protein
- self.entrez_to_uniprot[protein.entrez] = protein.uniprot_code
- self.gene_name_to_uniprot[protein.gene].add(protein.uniprot_code)
-
- def init_drugs(self):
- if len(self.drugs) == 0:
- print("Generating drug map...")
- for drug in models.Drug.objects.all():
- self.drugs[drug.drug_id] = drug
-
- def init_disorders(self):
- if len(self.disorders) == 0:
- print("Generating disorder map...")
- for disorder in models.Disorder.objects.all():
- self.disorders[disorder.mondo_id] = disorder
-
def import_proteins(self, update: bool):
proteins = dict()
gene_to_prots = defaultdict(lambda: set())
if update:
- self.init_proteins()
+ self.cache.init_proteins()
def add_protein(node):
- id = node['primaryDomainId'].split('.')[1]
+ id = to_id(node['primaryDomainId'])
name = node['geneName']
if len(node['synonyms']) > 0:
name = node['synonyms'][0]
@@ -100,13 +83,13 @@ class nedrex_importer:
proteins[id] = models.Protein(uniprot_code=id, protein_name=name, gene=node['geneName'])
def add_edges(edge):
- id = edge['sourceDomainId'].split('.')[1]
+ id = to_id(edge['sourceDomainId'])
protein = proteins[id]
- protein.entrez = edge['targetDomainId'].split('.')[1]
+ protein.entrez = to_id(edge['targetDomainId'])
gene_to_prots[protein.entrez].add(id)
def add_genes(node):
- id = node['primaryDomainId'].split('.')[1]
+ id = to_id(node['primaryDomainId'])
for prot_id in gene_to_prots[id]:
protein = proteins[prot_id]
try:
@@ -116,65 +99,177 @@ class nedrex_importer:
iter_node_collection('protein', add_protein)
iter_edge_collection('protein_encoded_by_gene', add_edges)
+
+ with_entrez = dict()
+ for ids in gene_to_prots.values():
+ for id in ids:
+ with_entrez[id] = proteins[id]
+ proteins = with_entrez
+
iter_node_collection('gene', add_genes)
# TODO test updating ideas
+
if update:
- (updates, creates) = identify_updates(proteins, self.proteins)
- models.Protein.objects.bulk_update(updates)
+ (updates, creates) = identify_updates(proteins, self.cache.proteins)
+ for u in updates:
+ u.save()
models.Protein.objects.bulk_create(creates)
for protein in creates:
- self.proteins[protein.uniprot_code] = protein
+ self.cache.proteins[protein.uniprot_code] = protein
else:
models.Protein.objects.bulk_create(proteins.values())
- self.proteins = proteins
- return len(self.proteins)
+ self.cache.proteins = proteins
+ return len(self.cache.proteins)
def import_drugs(self, update):
drugs = dict()
if update:
- self.init_drugs()
+ self.cache.init_drugs()
def add_drug(node):
- id = node['primaryDomainId'].split('.')[1]
+ id = to_id(node['primaryDomainId'])
drugs[id] = models.Drug(drug_id=id, name=node['displayName'], status=format_list(node['drugGroups']))
iter_node_collection('drug', add_drug)
# TODO test updating ideas
if update:
- (updates, creates) = identify_updates(drugs, self.drugs)
- models.Drug.objects.bulk_update(updates)
+ (updates, creates) = identify_updates(drugs, self.cache.drugs)
+ for u in updates:
+ u.save()
models.Drug.objects.bulk_create(creates)
for drug in creates:
- self.drugs[drug.drug_id] = drug
+ self.cache.drugs[drug.drug_id] = drug
else:
models.Drug.objects.bulk_create(drugs.values())
- self.drugs = drugs
+ self.cache.drugs = drugs
- return len(self.drugs)
+ return len(self.cache.drugs)
def import_disorders(self, update):
disorders = dict()
if update:
- self.init_disorders()
+ self.cache.init_disorders()
def add_disorder(node):
- id = node['primaryDomainId'].split('.')[1]
+ id = to_id(node['primaryDomainId'])
disorders[id] = models.Disorder(mondo_id=id, label=node['displayName'], icd10=format_list(node['icd10']))
iter_node_collection('disorder', add_disorder)
# TODO test updating ideas
if update:
- (updates, creates) = identify_updates(disorders, self.disorders)
- models.Disorder.objects.bulk_update(updates)
+ (updates, creates) = identify_updates(disorders, self.cache.disorders)
+ for u in updates:
+ u.save()
models.Disorder.objects.bulk_create(creates)
for disorder in creates:
- self.disorders[disorder.uniprot_code] = disorder
+ self.cache.disorders[disorder.mondo_id] = disorder
else:
models.Disorder.objects.bulk_create(disorders.values())
- self.disorders = disorders
+ self.cache.disorders = disorders
+
+ return len(self.cache.disorders)
+
+ def import_drug_target_interactions(self, dataset, update):
+ self.cache.init_drugs()
+ self.cache.init_proteins()
+
+ if update:
+ models.ProteinDrugInteraction.objects.filter(pdi_dataset=dataset).delete()
+
+ bulk = set()
- return len(self.disorders)
+ def add_dpi(edge):
+ try:
+ bulk.add(models.ProteinDrugInteraction(pdi_dataset=dataset,
+ drug=self.cache.get_drug_by_drugbank(
+ to_id(edge['sourceDomainId'])),
+ protein=self.cache.get_protein_by_uniprot(
+ to_id(edge['targetDomainId']))))
+ except KeyError:
+ pass
+ iter_edge_collection('drug_has_target', add_dpi)
+ models.ProteinDrugInteraction.objects.bulk_create(bulk)
+ return len(bulk)
+ def import_protein_protein_interactions(self, dataset, update):
+ self.cache.init_proteins()
+
+ if update:
+ models.ProteinProteinInteraction.objects.filter(ppi_dataset=dataset).delete()
+
+ bulk = list()
+
+ def iter_ppi(eval):
+ from python_nedrex import ppi
+ offset = 0
+ limit = 10000
+ while True:
+ result = ppi.ppis({"exp"}, skip=offset, limit=limit)
+ if not result:
+ return
+ for edge in result:
+ eval(edge)
+ offset += limit
+
+ def add_ppi(edge):
+ try:
+ bulk.append(models.ProteinProteinInteraction(ppi_dataset=dataset,
+ from_protein=self.cache.get_protein_by_uniprot(
+ to_id(edge['memberOne'])),
+ to_protein=self.cache.get_protein_by_uniprot(
+ to_id(edge['memberTwo']))))
+ except KeyError:
+ pass
+
+ iter_ppi(add_ppi)
+ models.ProteinProteinInteraction.objects.bulk_create(bulk)
+ return len(bulk)
+
+ def import_protein_disorder_associations(self, dataset, update):
+ self.cache.init_disorders()
+ self.cache.init_proteins()
+
+ if update:
+ models.ProteinDisorderAssociation.objects.filter(pdis_dataset=dataset).delete()
+
+ bulk = set()
+
+ def add_pdis(edge):
+ try:
+ disorder = self.cache.get_disorder_by_mondo(to_id(edge['targetDomainId']))
+ for protein in self.cache.get_proteins_by_entrez(to_id(edge['sourceDomainId'])):
+ bulk.add(models.ProteinDisorderAssociation(pdis_dataset=dataset,
+ protein=protein,
+ disorder=disorder, score=edge['score']))
+ except KeyError:
+ pass
+
+ iter_edge_collection('gene_associated_with_disorder', add_pdis)
+ models.ProteinDisorderAssociation.objects.bulk_create(bulk)
+ return len(bulk)
+
+ def import_drug_disorder_indications(self, dataset, update):
+ self.cache.init_disorders()
+ self.cache.init_drugs()
+
+ if update:
+ models.DrugDisorderIndication.objects.filter(drdi_dataset=dataset).delete()
+
+ bulk = set()
+
+ def add_drdis(edge):
+ try:
+ bulk.add(models.DrugDisorderIndication(drdi_dataset=dataset,
+ drug=self.cache.get_drug_by_drugbank(
+ to_id(edge['sourceDomainId'])),
+ disorder=self.cache.get_disorder_by_mondo(
+ to_id(edge['targetDomainId']))))
+ except KeyError:
+ pass
+
+ iter_edge_collection('drug_has_indication', add_drdis)
+ models.DrugDisorderIndication.objects.bulk_create(bulk)
+ return len(bulk)
diff --git a/drugstone/management/commands/make_graphs.py b/drugstone/management/commands/make_graphs.py
index 168c837..a6c3d81 100755
--- a/drugstone/management/commands/make_graphs.py
+++ b/drugstone/management/commands/make_graphs.py
@@ -80,7 +80,7 @@ def _internal_ppis(dataset_name: str) -> List[dict]:
def create_gt(params: Tuple[str, str]) -> None:
"""Fetches all required information to build a graph-tools file for given
PPI and PDI dataset names (params). Builds the graph-tools file and saves it in
- the data_drugstone/Networks folder.
+ the data/Networks folder.
Args:
params (Tuple[str, str]): Protein-protein-dataset name, Protein-drug-dataset name
@@ -176,7 +176,7 @@ def create_gt(params: Tuple[str, str]) -> None:
print("done with drug edges")
# save graph
- filename = f"./data_drugstone/Networks/internal_{ppi_dataset}_{pdi_dataset}.gt"
+ filename = f"./data/Networks/internal_{ppi_dataset}_{pdi_dataset}.gt"
g.save(filename)
print(f"Created file {filename}")
return
diff --git a/drugstone/management/commands/populate_db.py b/drugstone/management/commands/populate_db.py
index 00f3d7c..66d878f 100755
--- a/drugstone/management/commands/populate_db.py
+++ b/drugstone/management/commands/populate_db.py
@@ -1,28 +1,20 @@
from django.core.management.base import BaseCommand
-import pandas as pd
-from django.db import OperationalError, IntegrityError
+from django.db import OperationalError
-from drugstone.models import Protein, Drug, Tissue, ExpressionLevel, PPIDataset, PDIDataset, Disorder, PDisDataset, DrDiDataset
-from drugstone.models import ProteinProteinInteraction, ProteinDrugInteraction
+from drugstone.models import Protein, Drug, Tissue, ExpressionLevel, PPIDataset, PDIDataset, Disorder, PDisDataset, \
+ DrDiDataset, EnsemblGene
+from drugstone.models import ProteinProteinInteraction, ProteinDrugInteraction, ProteinDisorderAssociation, \
+ DrugDisorderIndication
from drugstone.management.includes.DataPopulator import DataPopulator
-from .import_from_nedrex import nedrex_importer
+from .import_from_nedrex import NedrexImporter
+from drugstone.management.includes.NodeCache import NodeCache
+from drugstone.management.includes import DatasetLoader
class DatabasePopulator:
- def __init__(self, data_dir,
- # protein_file,
- # drug_file,
- # protein_protein_interaction_file,
- # protein_drug_interaction_file,
- tissue_expression_file,
- ):
+ def __init__(self, data_dir):
self.data_dir = data_dir
- # self.protein_file = protein_file
- # self.drug_file = drug_file
- # self.ppi_file = protein_protein_interaction_file
- # self.pdi_file = protein_drug_interaction_file
- self.exp_file = tissue_expression_file
def delete_model(self, model):
from django.db import connection
@@ -32,31 +24,37 @@ class DatabasePopulator:
except OperationalError:
cursor.execute('DELETE FROM "{0}"'.format(model._meta.db_table))
+ def delete_all(self):
+ models = ['PPI', 'PDI', 'DrDi', 'Protein', 'Drug', 'Disorder', 'PDi', 'Expression', 'Tissue']
+ self.delete_models(models)
+
def delete_models(self, model_list):
for model_name in model_list:
print(f'Deleting {model_name} model ...')
if model_name == 'PPI':
+ self.delete_model(PPIDataset)
self.delete_model(ProteinProteinInteraction)
elif model_name == 'PDI':
+ self.delete_model(PDIDataset)
self.delete_model(ProteinDrugInteraction)
+ elif model_name == 'DrDi':
+ self.delete_model(DrDiDataset)
+ self.delete_model(DrugDisorderIndication)
elif model_name == 'Protein':
self.delete_model(Protein)
+ self.delete_model(EnsemblGene)
elif model_name == 'Drug':
self.delete_model(Drug)
elif model_name == 'Disorder':
self.delete_model(Disorder)
- elif model_name == 'PDiAssociations':
+ elif model_name == 'PDi':
self.delete_model(PDisDataset)
+ self.delete_model(ProteinDisorderAssociation)
+ elif model_name == 'Expression':
+ self.delete_model(ExpressionLevel)
elif model_name == 'Tissue':
self.delete_model(Tissue)
- elif model_name == 'PPIDataset':
- self.delete_model(PPIDataset)
- elif model_name == 'PDIDataset':
- self.delete_model(PDIDataset)
- elif model_name == 'DrDiDataset':
- self.delete_model(DrDiDataset)
-
class Command(BaseCommand):
@@ -65,104 +63,126 @@ class Command(BaseCommand):
# dataset directory
parser.add_argument('-dd', '--data_dir', type=str, help='Dataset directory path')
parser.add_argument('-dm', '--delete_model', type=str, help='Delete model(s)')
+ parser.add_argument('-c', '--clear', action='store_true', help='Delete all models')
+ parser.add_argument('-a', '--all', action='store_true', help='Populate all tables')
+ parser.add_argument('-u', '--update', action='store_true', help='Execute database update for selected tables')
parser.add_argument('-p', '--proteins', action='store_true', help='Populate Proteins')
parser.add_argument('-di', '--disorders', action='store_true', help='Populate Disorders')
parser.add_argument('-dr', '--drugs', action='store_true', help='Drug file name')
- parser.add_argument('-exp', '--exp_file', type=str, help='Tissue expression file (.gct without first 2 lines)')
+ parser.add_argument('-exp', '--exp', action='store_true',
+ help='Tissue expression file (.gct without first 2 lines)')
- parser.add_argument('-pp', '--protein_protein', type=str, help='Populate Protein-Protein Interactions')
- parser.add_argument('-pdr', '--protein_drug', type=str, help='Populate Protein-Drug Interactions')
- parser.add_argument('-pdi', '--protein_disorder', type=str, help='Populate Protein-Disorder Associations')
- parser.add_argument('-ddi', '--drug_disorder', type=str, help='Populate Drug-Disorder Indications')
+ parser.add_argument('-pp', '--protein_protein', action='store_true',
+ help='Populate Protein-Protein Interactions')
+ parser.add_argument('-pdr', '--protein_drug', action='store_true', help='Populate Protein-Drug Interactions')
+ parser.add_argument('-pdi', '--protein_disorder', action='store_true',
+ help='Populate Protein-Disorder Associations')
+ parser.add_argument('-ddi', '--drug_disorder', action='store_true', help='Populate Drug-Disorder Indications')
def handle(self, *args, **kwargs):
-
+ nedrex_api_url = "http://82.148.225.92:8123/"
data_dir = kwargs['data_dir']
- exp_file = kwargs['exp_file']
-
- # p = kwargs['proteins']
- # pp = kwargs['protein_protein']
- # pd = kwargs['protein_drug']
+ db_populator = DatabasePopulator(data_dir=data_dir)
- db_populator = DatabasePopulator(data_dir=data_dir,
- # protein_file=protein_file,
- # drug_file=drug_file,
- # protein_protein_interaction_file=ppi_file,
- # protein_drug_interaction_file=pdi_file,
- tissue_expression_file=exp_file,
- )
-
- importer = nedrex_importer("http://82.148.225.92:8123/")
+ if kwargs['clear']:
+ db_populator.delete_all()
if kwargs['delete_model'] is not None:
model_list = kwargs['delete_model'].split(',')
db_populator.delete_models(model_list)
- return
- populator = DataPopulator()
+ cache = NodeCache()
+ update = True if kwargs['update'] else False
+ importer = NedrexImporter(nedrex_api_url, cache)
+ populator = DataPopulator(cache)
+
+ if kwargs['all']:
+ kwargs['drugs'] = True
+ kwargs['disorders'] = True
+ kwargs['proteins'] = True
+ kwargs['exp'] = True
+ kwargs['protein_protein'] = True
+ kwargs['protein_drug'] = True
+ kwargs['protein_disorder'] = True
+ kwargs['drug_disorder'] = True
if kwargs['drugs']:
print('Populating Drugs...')
- # n = DataPopulator.populate_drugs(populator)
- n = nedrex_importer.import_drugs(importer,False)
+ n = NedrexImporter.import_drugs(importer, update)
print(f'Populated {n} Drugs.')
-
- if kwargs['exp_file'] is not None:
- print('Populating Expressions...')
- n = DataPopulator.populate_expessions(populator)
- print(f'Populated {n} Expressions.')
+ if kwargs['disorders']:
+ print('Populating Disorders...')
+ n = NedrexImporter.import_disorders(importer, update)
+ print(f'Populated {n} Disorders.')
if kwargs['proteins']:
print('Populating Proteins...')
-
- n = nedrex_importer.import_proteins(importer, False)
- # n = DataPopulator.populate_proteins(populator)
+ n = NedrexImporter.import_proteins(importer, update)
print(f'Populated {n} Proteins.')
-
- # print('Populating ENSG IDs...')
- # n = DataPopulator.populate_ensg(populator)
- # print(f'Populated {n} ENSG IDs.')
+ print('Populating ENSG IDs...')
+ n = DataPopulator.populate_ensg(populator,update)
+ print(f'Populated {n} ENSG IDs.')
- if kwargs['disorders']:
- print('Populating Disorders...')
- n = nedrex_importer.import_disorders(importer, False)
- # n = DataPopulator.populate_disorders(populator)
- print(f'Populated {n} Disorders.')
+ if kwargs['exp']:
+ print('Populating Expressions...')
+ n = DataPopulator.populate_expressions(populator, update)
+ print(f'Populated {n} Expressions.')
- if kwargs['protein_protein'] is not None:
+ if kwargs['protein_protein']:
+ print('Importing PPIs from NeDRexDB...')
+ n = NedrexImporter.import_protein_protein_interactions(importer,
+ DatasetLoader.get_ppi_nedrex(nedrex_api_url),
+ update)
+ print(f'Imported {n} PPIs from NeDRexDB')
print('Populating PPIs from STRING...')
- n = DataPopulator.populate_ppi_string(populator)
+ n = DataPopulator.populate_ppi_string(populator, DatasetLoader.get_ppi_string(), update)
print(f'Populated {n} PPIs from STRING.')
print('Populating PPIs from APID...')
- n = DataPopulator.populate_ppi_apid(populator)
+ n = DataPopulator.populate_ppi_apid(populator, DatasetLoader.get_ppi_apid(), update)
print(f'Populated {n} PPIs from APID.')
print('Populating PPIs from BioGRID...')
- n = DataPopulator.populate_ppi_biogrid(populator)
+ n = DataPopulator.populate_ppi_biogrid(populator, DatasetLoader.get_ppi_biogrid(), update)
print(f'Populated {n} PPIs from BioGRID.')
- if kwargs['protein_drug'] is not None:
+ if kwargs['protein_drug']:
+ print('Importing PDIs from NeDRexDB...')
+ n = NedrexImporter.import_drug_target_interactions(importer, DatasetLoader.get_drug_target_nedrex(nedrex_api_url), update)
+ print(f'Imported {n} PDIs from NeDRexDB')
+
print('Populating PDIs from Chembl...')
- n = DataPopulator.populate_pdi_chembl(populator)
+ n = DataPopulator.populate_pdi_chembl(populator,DatasetLoader.get_drug_target_chembl(), update)
print(f'Populated {n} PDIs from Chembl.')
print('Populating PDIs from DGIdb...')
- n = DataPopulator.populate_pdi_dgidb(populator)
+ n = DataPopulator.populate_pdi_dgidb(populator, DatasetLoader.get_drug_target_dgidb(), update)
print(f'Populated {n} PDIs from DGIdb.')
print('Populating PDIs from DrugBank...')
- n = DataPopulator.populate_pdi_drugbank(populator)
+ n = DataPopulator.populate_pdi_drugbank(populator, DatasetLoader.get_drug_target_drugbank(), update)
print(f'Populated {n} PDIs from DrugBank.')
- if kwargs['protein_disorder'] is not None:
+
+ if kwargs['protein_disorder']:
+ print('Importing PDis from NeDRexDB...')
+ n = NedrexImporter.import_protein_disorder_associations(importer,
+ DatasetLoader.get_protein_disorder_nedrex(nedrex_api_url),
+ update)
+ print(f'Imported {n} PDis from NeDRexDB')
print('Populating PDis associations from DisGeNET...')
- n=DataPopulator.populate_pdis_disgenet(populator)
+ n = DataPopulator.populate_pdis_disgenet(populator, DatasetLoader.get_disorder_protein_disgenet(), update)
print(f'Populated {n} PDis associations from DisGeNET.')
- if kwargs['drug_disorder'] is not None:
+
+ if kwargs['drug_disorder']:
+ print('Importing DrDis from NeDRexDB...')
+ n = NedrexImporter.import_drug_disorder_indications(importer,
+ DatasetLoader.get_drug_disorder_nedrex(nedrex_api_url),
+ update)
+ print(f'Imported {n} DrDis from NeDRexDB')
print('Populating DrDi indications from DrugBank...')
- n=DataPopulator.populate_drdis_drugbank(populator)
+ n = DataPopulator.populate_drdis_drugbank(populator, DatasetLoader.get_drug_disorder_drugbank(), update)
print(f'Populated {n} DrDi associations from DrugBank.')
diff --git a/drugstone/management/commands/test.py b/drugstone/management/commands/test.py
index 5afbea2..4d01dd4 100644
--- a/drugstone/management/commands/test.py
+++ b/drugstone/management/commands/test.py
@@ -1,5 +1,6 @@
import python_nedrex as nedrex
from python_nedrex.core import get_nodes, get_edges, get_api_key
+from python_nedrex.static import get_metadata
def iter_node_collection(coll_name, eval):
offset = 0
@@ -25,9 +26,26 @@ def iter_edge_collection(coll_name, eval):
offset += limit
+def iter_ppi(eval):
+ from python_nedrex import ppi
+ offset = 0
+ limit = 1000
+ while True:
+ result = ppi.ppis({"exp"},skip = offset, limit=limit)
+ if not result:
+ return
+ for edge in result:
+ eval(edge)
+ offset += limit
+
base_url = "http://82.148.225.92:8123/"
nedrex.config.set_url_base(base_url)
api_key = get_api_key(accept_eula=True)
nedrex.config.set_api_key(api_key)
+print(f'Nodes: {nedrex.core.get_node_types()}')
+print(f'Edges: {nedrex.core.get_edge_types()}')
+print(f'{get_metadata()}')
+
-iter_edge_collection("gene_expressed_in_tissue", lambda node: {print(node)})
\ No newline at end of file
+iter_ppi(lambda node: print(node))
+# iter_edge_collection("gene_expressed_in_tissue", lambda node: {print(node)})
\ No newline at end of file
diff --git a/drugstone/management/includes/DataLoader.py b/drugstone/management/includes/DataLoader.py
index 8fa0d2a..ca31b56 100755
--- a/drugstone/management/includes/DataLoader.py
+++ b/drugstone/management/includes/DataLoader.py
@@ -3,14 +3,14 @@ import json
class DataLoader:
- PATH_PROTEINS = 'data_drugstone/Proteins/'
- PATH_DRUGS = 'data_drugstone/Drugs/'
- PATH_EXPR = 'data_drugstone/'
- PATH_DISORDERS = 'data_drugstone/Disorders/'
- PATH_PDI = 'data_drugstone/PDI/'
- PATH_PPI = 'data_drugstone/PPI/'
- PATH_PDi = 'data_drugstone/PDi/'
- PATH_DDi = 'data_drugstone/DrDi/'
+ PATH_PROTEINS = 'data/Proteins/'
+ PATH_DRUGS = 'data/Drugs/'
+ PATH_EXPR = 'data/Expression/'
+ PATH_DISORDERS = 'data/Disorders/'
+ PATH_PDI = 'data/PDI/'
+ PATH_PPI = 'data/PPI/'
+ PATH_PDi = 'data/PDi/'
+ PATH_DDi = 'data/DrDi/'
# Proteins
PROTEINS_COVEX = 'protein_list.csv'
@@ -230,7 +230,7 @@ class DataLoader:
Returns:
pd.DataFrame: columns "protein_name", "disorder_name" and "score"
"""
- return pd.read_csv(f'{DataLoader.PATH_PDi}{DataLoader.PDi_DISGENET}', sep='\t')
+ return pd.read_csv(f'{DataLoader.PATH_PDi}{DataLoader.PDi_DISGENET}', sep='\t', dtype={'disorder_name':str, 'protein_name':str, 'score':float})
@staticmethod
def load_drdis_drugbank() -> pd.DataFrame:
@@ -239,7 +239,7 @@ class DataLoader:
Returns:
pd.DataFrame: columns "drugbank_id" and "mondo_id"
"""
- return pd.read_csv(f'{DataLoader.PATH_DDi}{DataLoader.DDi_DRUGBANK}', sep='\t')
+ return pd.read_csv(f'{DataLoader.PATH_DDi}{DataLoader.DDi_DRUGBANK}', sep='\t', dtype={'drugbank_id':str, 'mondo_id':str})
@staticmethod
def load_pdi_dgidb() -> pd.DataFrame:
diff --git a/drugstone/management/includes/DataPopulator.py b/drugstone/management/includes/DataPopulator.py
index 4056232..5858202 100755
--- a/drugstone/management/includes/DataPopulator.py
+++ b/drugstone/management/includes/DataPopulator.py
@@ -1,88 +1,18 @@
-from collections import defaultdict
-
from drugstone.management.includes.DataLoader import DataLoader
import drugstone.models as models
+from drugstone.management.includes.NodeCache import NodeCache
class DataPopulator:
- proteins = dict()
- uniprot_to_ensembl = dict()
- gene_name_to_ensembl = defaultdict(lambda: set())
- disorders = dict()
- drugs = dict()
-
- def init_proteins(self):
- if len(self.proteins) == 0:
- print("Generating protein maps...")
- for protein in models.Protein.objects.all():
- self.proteins[protein.entrez]=protein
- self.uniprot_to_ensembl[protein.uniprot_code] = protein.entrez
- self.gene_name_to_ensembl[protein.gene].add(protein.entrez)
-
- def init_drugs(self):
- if len(self.drugs)== 0:
- print("Generating drug map...")
- for drug in models.Drug.objects.all():
- self.drugs[drug.drug_id]=drug
-
- def init_disorders(self):
- if len(self.disorders) == 0:
- print("Generating disorder map...")
- for disorder in models.Disorder.objects.all():
- self.disorders[disorder.mondo_id]=disorder
-
- # def populate_proteins(self) -> int:
- # """ Populates the Protein table in the django database.
- # Handles loading the data and passing it to the django database
- #
- # Returns:
- # int: Count of how many proteins were added
- # """
- # df = DataLoader.load_proteins()
- # for _, row in df.iterrows():
- # self.proteins[row['entrez_id']] = models.Protein(
- # uniprot_code=row['protein_ac'],
- # gene=row['gene_name'],
- # entrez=row['entrez_id'],
- # protein_name=row['protein_name'])
- # self.uniprot_to_ensembl[row['protein_ac']] = row['entrez_id']
- # self.gene_name_to_ensembl[row['gene_name']].add(row['entrez_id'])
- #
- # models.Protein.objects.bulk_create(self.proteins.values())
- # return len(self.proteins)
- #
- # def populate_disorders(self) -> int:
- # """ Populates the Disorder table in the django database.
- # Handles loading the data and passing it to the django database
- #
- # Returns:
- # int: Count of how many disorders were added
- # """
- # df = DataLoader.load_disorders()
- # for _, row in df.iterrows():
- # self.disorders[row['mondo_id']] = models.Disorder(
- # mondo_id=row['mondo_id'],
- # label=row['label'],
- # icd10=row['icd10']
- # )
- # models.Disorder.objects.bulk_create(self.disorders.values())
- # return len(self.disorders)
- #
- # def populate_drugs(self):
- # df = DataLoader.load_drugs()
- # for _, row in df.iterrows():
- # drug_id = row['drug_id']
- # drug_name = row['drug_name']
- # drug_status = row['drug_status']
- # self.drugs[drug_id] = models.Drug(
- # drug_id=drug_id,
- # name=drug_name,
- # status=drug_status)
- # models.Drug.objects.bulk_create(self.drugs.values())
- # return len(self.drugs)
-
- def populate_expessions(self):
- self.init_proteins()
+
+ def __init__(self, cache: NodeCache):
+ self.cache = cache
+
+ def populate_expressions(self, update):
+ if update:
+ models.ExpressionLevel.objects.all().delete()
+
+ self.cache.init_proteins()
df = DataLoader.load_expressions()
tissues_models = dict()
@@ -94,28 +24,34 @@ class DataPopulator:
tissues_models[tissue_name] = tissue_model
proteins_linked = 0
- unique = set()
- bulk = list()
+ bulk = set()
+ uniq = set()
+ size = 0
for _, row in df.iterrows():
gene_name = row['Description']
- for protein_id in self.gene_name_to_ensembl[gene_name]:
- protein_model = self.proteins[protein_id]
+ for protein_model in self.cache.get_proteins_by_gene(gene_name):
proteins_linked += 1
for tissue_name, tissue_model in tissues_models.items():
- id = f"{tissue_name}_{protein_id}"
- if id in unique:
+ expr = models.ExpressionLevel(protein=protein_model,
+ tissue=tissue_model,
+ expression_level=row[tissue_name])
+ id = hash(expr)
+ if id in uniq:
continue
- unique.add(id)
- bulk.append(models.ExpressionLevel(protein=protein_model,
- tissue=tissue_model,
- expression_level=row[tissue_name]))
+ uniq.add(id)
+ bulk.add(expr)
+ if len(bulk) > 100000:
+ models.ExpressionLevel.objects.bulk_create(bulk)
+ size += len(bulk)
+ bulk = set()
+
models.ExpressionLevel.objects.bulk_create(bulk)
- return len(bulk)
+ return size + len(bulk)
- def populate_ensg(self) -> int:
+ def populate_ensg(self,update) -> int:
""" Populates the Ensembl-Gene table in the django database.
Also maps the added ensg entries to the corresponding proteins.
Handles loading the data and passing it to the django database
@@ -123,76 +59,78 @@ class DataPopulator:
Returns:
int: Count of how many ensg-protein relations were added
"""
- self.init_proteins()
+ if update:
+ models.EnsemblGene.objects.all().delete()
+ self.cache.init_proteins()
data = DataLoader.load_ensg()
bulk = list()
+
for entrez, ensg_list in data.items():
- protein = self.proteins[entrez]
- for ensg in ensg_list:
- bulk.append(models.EnsemblGene(name=ensg, protein=protein))
+ proteins = self.cache.get_proteins_by_entrez(entrez)
+ for protein in proteins:
+ for ensg in ensg_list:
+ bulk.append(models.EnsemblGene(name=ensg, protein=protein))
models.EnsemblGene.objects.bulk_create(bulk)
return len(bulk)
- def populate_ppi_string(self) -> int:
+ def populate_ppi_string(self, dataset, update) -> int:
""" Populates the Protein-Protein-Interactions from STRINGdb
Handles loading the data and passing it to the django database
Returns:
int: Count of how many interactions were added
"""
- self.init_proteins()
+ self.cache.init_proteins()
+ if update:
+ models.ProteinProteinInteraction.objects.filter(ppi_dataset=dataset).delete()
+
df = DataLoader.load_ppi_string()
- dataset, _ = models.PPIDataset.objects.get_or_create(
- name='STRING',
- link='https://string-db.org/',
- version='11.0'
- )
bulk = list()
for _, row in df.iterrows():
try:
# try fetching proteins
- protein_a = self.proteins[row['entrez_a']]
- protein_b = self.proteins[row['entrez_b']]
+ proteins_a = self.cache.get_proteins_by_entrez(row['entrez_a'])
+ proteins_b = self.cache.get_proteins_by_entrez(row['entrez_b'])
except KeyError:
# continue if not found
continue
- try:
- bulk.append(models.ProteinProteinInteraction(
- ppi_dataset=dataset,
- from_protein=protein_a,
- to_protein=protein_b
- ))
- except models.ValidationError:
- # duplicate
- continue
+ for protein_a in proteins_a:
+ for protein_b in proteins_b:
+ try:
+ bulk.append(models.ProteinProteinInteraction(
+ ppi_dataset=dataset,
+ from_protein=protein_a,
+ to_protein=protein_b
+ ))
+ except models.ValidationError:
+ # duplicate
+ continue
models.ProteinProteinInteraction.objects.bulk_create(bulk)
return len(bulk)
- def populate_ppi_apid(self) -> int:
+ def populate_ppi_apid(self, dataset, update) -> int:
""" Populates the Protein-Protein-Interactions from Apid
Handles loading the data and passing it to the django database
Returns:
int: Count of how many interactions were added
"""
- self.init_proteins()
+ self.cache.init_proteins()
+
+ if update:
+ models.ProteinProteinInteraction.objects.filter(ppi_dataset=dataset).delete()
df = DataLoader.load_ppi_apid()
- dataset, _ = models.PPIDataset.objects.get_or_create(
- name='APID',
- link='http://cicblade.dep.usal.es:8080/APID/',
- version='January 2019'
- )
- bulk = list()
+ bulk = set()
for _, row in df.iterrows():
try:
# try fetching proteins
- protein_a = self.proteins[self.uniprot_to_ensembl[row['from_protein_ac']]]
- protein_b = self.proteins[self.uniprot_to_ensembl[row['to_protein_ac']]]
+ protein_a = self.cache.get_protein_by_uniprot(row['from_protein_ac'])
+ protein_b = self.cache.get_protein_by_uniprot(row['to_protein_ac'])
except KeyError:
# continue if not found
continue
try:
- bulk.append(models.ProteinProteinInteraction(
+ bulk.add(models.ProteinProteinInteraction(
ppi_dataset=dataset,
from_protein=protein_a,
to_protein=protein_b
@@ -202,71 +140,69 @@ class DataPopulator:
models.ProteinProteinInteraction.objects.bulk_create(bulk)
return len(bulk)
- def populate_ppi_biogrid(self) -> int:
+ def populate_ppi_biogrid(self,dataset, update) -> int:
""" Populates the Protein-Protein-Interactions from BioGRID
Handles loading the data and passing it to the django database
Returns:
int: Count of how many interactions were added
"""
- self.init_proteins()
+ self.cache.init_proteins()
+
+ if update:
+ models.ProteinProteinInteraction.objects.filter(ppi_dataset=dataset).delete()
df = DataLoader.load_ppi_biogrid()
- dataset, _ = models.PPIDataset.objects.get_or_create(
- name='BioGRID',
- link='https://thebiogrid.org/',
- version='4.0'
- )
bulk = list()
for _, row in df.iterrows():
try:
# try fetching proteins
- protein_a = self.proteins[row['entrez_a']]
- protein_b = self.proteins[row['entrez_b']]
+ proteins_a = self.cache.get_proteins_by_entrez(row['entrez_a'])
+ proteins_b = self.cache.get_proteins_by_entrez(row['entrez_b'])
except KeyError:
# TODO update error
# continue if not found
continue
- try:
- bulk.append(models.ProteinProteinInteraction(
- ppi_dataset=dataset,
- from_protein=protein_a,
- to_protein=protein_b
- ))
- except models.ValidationError:
- # duplicate
- continue
+ for protein_a in proteins_a:
+ for protein_b in proteins_b:
+ try:
+ bulk.append(models.ProteinProteinInteraction(
+ ppi_dataset=dataset,
+ from_protein=protein_a,
+ to_protein=protein_b
+ ))
+ except models.ValidationError:
+ # duplicate
+ continue
models.ProteinProteinInteraction.objects.bulk_create(bulk)
return len(bulk)
- def populate_pdi_chembl(self) -> int:
+ def populate_pdi_chembl(self,dataset, update) -> int:
""" Populates the Protein-Drug-Interactions from Chembl
Handles Loading the data and passing it to the django database
Returns:
int: Count of how many interactions were added
"""
- self.init_proteins()
- self.init_drugs()
+ self.cache.init_proteins()
+ self.cache.init_drugs()
+
+ if update:
+ models.ProteinDrugInteraction.objects.filter(pdi_dataset=dataset).delete()
df = DataLoader.load_pdi_chembl()
- dataset, _ = models.PDIDataset.objects.get_or_create(
- name='ChEMBL',
- link='https://www.ebi.ac.uk/chembl/',
- version='27',
- )
- bulk = list()
+ bulk = set()
for _, row in df.iterrows():
try:
- protein = self.proteins[self.uniprot_to_ensembl[row['protein_ac']]]
+ protein = self.cache.get_protein_by_uniprot(row['protein_ac'])
except KeyError:
# continue if not found
continue
try:
# try fetching drug
- drug = self.drugs[row['drug_id']]
+ drug = self.cache.get_drug_by_drugbank(row['drug_id'])
except KeyError:
# continue if not found
continue
- bulk.append(models.ProteinDrugInteraction(
+ bulk.add(models.ProteinDrugInteraction(
pdi_dataset=dataset,
protein=protein,
drug=drug
@@ -274,36 +210,35 @@ class DataPopulator:
models.ProteinDrugInteraction.objects.bulk_create(bulk)
return len(bulk)
- def populate_pdis_disgenet(self,) -> int:
+ def populate_pdis_disgenet(self, dataset, update) -> int:
""" Populates the Protein-Disorder-Interactions from DisGeNET
Handles Loading the data and passing it to the django database
Returns:
int: Count of how many interactions were added
"""
- self.init_proteins()
- self.init_disorders()
+ self.cache.init_proteins()
+ self.cache.init_disorders()
+
+
+ if update:
+ models.ProteinDisorderAssociation.objects.filter(pdis_dataset=dataset).delete()
df = DataLoader.load_pdis_disgenet()
- dataset, _ = models.PDisDataset.objects.get_or_create(
- name='DisGeNET',
- link='https://www.disgenet.org/home/',
- version='6.0',
- )
- bulk = list()
+ bulk = set()
for _, row in df.iterrows():
try:
# try fetching protein
- protein = self.proteins[self.uniprot_to_ensembl[row['protein_name']]]
+ protein = self.cache.get_protein_by_uniprot(row['protein_name'])
except KeyError:
# continue if not found
continue
try:
- # try fetching drug
- disorder = self.disorders[str(int(row['disorder_name']))]
+ # try fetching disorder
+ disorder = self.cache.get_disorder_by_mondo(row['disorder_name'])
except KeyError:
# continue if not found
continue
- bulk.append(models.ProteinDisorderAssociation(
+ bulk.add(models.ProteinDisorderAssociation(
pdis_dataset=dataset,
protein=protein,
disorder=disorder,
@@ -312,36 +247,34 @@ class DataPopulator:
models.ProteinDisorderAssociation.objects.bulk_create(bulk)
return len(bulk)
- def populate_drdis_drugbank(self) -> int:
+ def populate_drdis_drugbank(self, dataset, update) -> int:
""" Populates the Drug-Disorder-Indications from DrugBank
Handles Loading the data and passing it to the django database
Returns:
int: Count of how many edges were added
"""
- self.init_drugs()
- self.init_disorders()
+ self.cache.init_drugs()
+ self.cache.init_disorders()
+ if update:
+ models.DrugDisorderIndication.objects.filter(drdi_dataset=dataset).delete()
+
df = DataLoader.load_drdis_drugbank()
- dataset, _ = models.DrDiDataset.objects.get_or_create(
- name='DrugBank',
- link='https://go.drugbank.com/',
- version='5.1.8',
- )
- bulk = list()
+ bulk = set()
for _, row in df.iterrows():
try:
# try fetching protein
- drug = self.drugs[row['drugbank_id']]
+ drug = self.cache.get_drug_by_drugbank(row['drugbank_id'])
except KeyError:
# continue if not found
continue
try:
# try fetching drug
- disorder = self.disorders[str(int(row['mondo_id']))]
+ disorder = self.cache.get_disorder_by_mondo(row['mondo_id'])
except KeyError:
# continue if not found
continue
- bulk.append(models.DrugDisorderIndication(
+ bulk.add(models.DrugDisorderIndication(
drdi_dataset=dataset,
drug=drug,
disorder=disorder,
@@ -349,76 +282,75 @@ class DataPopulator:
models.DrugDisorderIndication.objects.bulk_create(bulk)
return len(bulk)
- def populate_pdi_dgidb(self) -> int:
+ def populate_pdi_dgidb(self,dataset, update) -> int:
""" Populates the Protein-Drug-Interactions from DGIdb
Handles Loading the data and passing it to the django database
Returns:
int: Count of how many interactions were added
"""
- self.init_proteins()
- self.init_drugs()
+ self.cache.init_proteins()
+ self.cache.init_drugs()
+
+ if update:
+ models.ProteinDrugInteraction.objects.filter(pdi_dataset=dataset).delete()
df = DataLoader.load_pdi_dgidb()
- dataset, _ = models.PDIDataset.objects.get_or_create(
- name='DGIdb',
- link='https://www.dgidb.org/',
- version='4.2.0'
- )
- bulk = list()
+ bulk = set()
for _, row in df.iterrows():
try:
# try fetching protein
- protein = self.proteins[row['entrez_id']]
+ proteins = self.cache.get_proteins_by_entrez(row['entrez_id'])
except KeyError:
# continue if not found
continue
try:
# try fetching drug
- drug = self.drugs[row['drug_id']]
+ drug = self.cache.get_drug_by_drugbank(row['drug_id'])
except KeyError:
# continue if not found
continue
- bulk.append(models.ProteinDrugInteraction(
- pdi_dataset=dataset,
- protein=protein,
- drug=drug
- ))
+ for protein in proteins:
+ bulk.add(models.ProteinDrugInteraction(
+ pdi_dataset=dataset,
+ protein=protein,
+ drug=drug
+ ))
models.ProteinDrugInteraction.objects.bulk_create(bulk)
return len(bulk)
- def populate_pdi_drugbank(self) -> int:
+ def populate_pdi_drugbank(self,dataset, update) -> int:
""" Populates the Protein-Drug-Interactions from Drugbank
Handles Loading the data and passing it to the django database
Returns:
int: Count of how many interactions were added
"""
- self.init_proteins()
- self.init_drugs()
+ self.cache.init_proteins()
+ self.cache.init_drugs()
+
+
+ if update:
+ models.ProteinDrugInteraction.objects.filter(pdi_dataset=dataset).delete()
df = DataLoader.load_pdi_drugbank()
- dataset, _ = models.PDIDataset.objects.get_or_create(
- name='DrugBank',
- link='https://go.drugbank.com/',
- version='5.1.7'
- )
- bulk = list()
+ bulk = set()
for _, row in df.iterrows():
try:
# try fetching protein
- protein = self.proteins[row['entrez_id']]
+ proteins = self.cache.get_proteins_by_entrez(row['entrez_id'])
except KeyError:
# continue if not found
continue
try:
# try fetching drug
- drug = self.drugs[row['drug_id']]
+ drug = self.cache.get_drug_by_drugbank(row['drug_id'])
except KeyError:
# continue if not found
continue
- bulk.append(models.ProteinDrugInteraction(
- pdi_dataset=dataset,
- protein=protein,
- drug=drug
- ))
+ for protein in proteins:
+ bulk.add(models.ProteinDrugInteraction(
+ pdi_dataset=dataset,
+ protein=protein,
+ drug=drug
+ ))
models.ProteinDrugInteraction.objects.bulk_create(bulk)
return len(bulk)
diff --git a/drugstone/management/includes/DatasetLoader.py b/drugstone/management/includes/DatasetLoader.py
new file mode 100644
index 0000000..f608ed1
--- /dev/null
+++ b/drugstone/management/includes/DatasetLoader.py
@@ -0,0 +1,99 @@
+from drugstone import models
+from python_nedrex.static import get_metadata
+
+def get_ppi_string():
+ dataset, _ = models.PPIDataset.objects.get_or_create(
+ name='STRING',
+ link='https://string-db.org/',
+ version='11.0'
+ )
+ return dataset
+
+def get_ppi_apid():
+ dataset, _ = models.PPIDataset.objects.get_or_create(
+ name='APID',
+ link='http://cicblade.dep.usal.es:8080/APID/',
+ version='January 2019'
+ )
+ return dataset
+
+def get_ppi_biogrid():
+ dataset, _ = models.PPIDataset.objects.get_or_create(
+ name='BioGRID',
+ link='https://thebiogrid.org/',
+ version='4.0'
+ )
+ return dataset
+
+def get_drug_target_nedrex(url):
+ dataset, _ = models.PDIDataset.objects.get_or_create(
+ name='NeDRex',
+ link=url,
+ version=get_metadata()['version'],
+ )
+ return dataset
+
+def get_ppi_nedrex(url):
+ dataset, _ = models.PPIDataset.objects.get_or_create(
+ name='NeDRex',
+ link=url,
+ version=get_metadata()['version'],
+ )
+ return dataset
+
+def get_protein_disorder_nedrex(url):
+ dataset, _ = models.PDisDataset.objects.get_or_create(
+ name='NeDRex',
+ link=url,
+ version=get_metadata()['version'],
+ )
+ return dataset
+
+def get_drug_disorder_nedrex(url):
+ dataset, _ = models.DrDiDataset.objects.get_or_create(
+ name='NeDRex',
+ link=url,
+ version=get_metadata()['version'],
+ )
+ return dataset
+
+def get_drug_target_chembl():
+ dataset, _ = models.PDIDataset.objects.get_or_create(
+ name='ChEMBL',
+ link='https://www.ebi.ac.uk/chembl/',
+ version='27',
+ )
+ return dataset
+
+def get_drug_target_dgidb():
+ dataset, _ = models.PDIDataset.objects.get_or_create(
+ name='DGIdb',
+ link='https://www.dgidb.org/',
+ version='4.2.0'
+ )
+ return dataset
+
+def get_drug_target_drugbank():
+ dataset, _ = models.PDIDataset.objects.get_or_create(
+ name='DrugBank',
+ link='https://go.drugbank.com/',
+ version='5.1.7'
+ )
+ return dataset
+
+def get_disorder_protein_disgenet():
+ dataset, _ = models.PDisDataset.objects.get_or_create(
+ name='DisGeNET',
+ link='https://www.disgenet.org/home/',
+ version='6.0',
+ )
+ return dataset
+
+
+def get_drug_disorder_drugbank():
+ dataset, _ = models.DrDiDataset.objects.get_or_create(
+ name='DrugBank',
+ link='https://go.drugbank.com/',
+ version='5.1.8',
+ )
+ return dataset
diff --git a/drugstone/management/includes/NodeCache.py b/drugstone/management/includes/NodeCache.py
new file mode 100644
index 0000000..7f9491c
--- /dev/null
+++ b/drugstone/management/includes/NodeCache.py
@@ -0,0 +1,60 @@
+from collections import defaultdict
+import drugstone.models as models
+
+
+class NodeCache:
+
+ proteins = dict()
+ entrez_to_uniprot = defaultdict(lambda: set())
+ gene_name_to_uniprot = defaultdict(lambda: set())
+ disorders = dict()
+ drugs = dict()
+
+ def init_protein_maps(self):
+ print("Generating protein id maps...")
+ for protein in self.proteins.values():
+ self.entrez_to_uniprot[protein.entrez].add(protein.uniprot_code)
+ self.gene_name_to_uniprot[protein.gene].add(protein.uniprot_code)
+
+ def init_proteins(self):
+ if len(self.proteins) == 0:
+ print("Generating protein maps...")
+ for protein in models.Protein.objects.all():
+ self.proteins[protein.uniprot_code] = protein
+ if len(self.proteins) > 0 and (len(self.entrez_to_uniprot) == 0 or len(self.gene_name_to_uniprot) == 0):
+ self.init_protein_maps()
+
+
+ def init_drugs(self):
+ if len(self.drugs) == 0:
+ print("Generating drug map...")
+ for drug in models.Drug.objects.all():
+ self.drugs[drug.drug_id] = drug
+
+ def init_disorders(self):
+ if len(self.disorders) == 0:
+ print("Generating disorder map...")
+ for disorder in models.Disorder.objects.all():
+ self.disorders[disorder.mondo_id] = disorder
+
+
+ def get_protein_by_uniprot(self,uniprot_id):
+ return self.proteins[uniprot_id]
+
+ def get_proteins_by_entrez(self,entrez_id):
+ out = list()
+ for g in self.entrez_to_uniprot[entrez_id]:
+ out.append(self.proteins[g])
+ return out
+
+ def get_proteins_by_gene(self, gene_name):
+ out = list()
+ for g in self.gene_name_to_uniprot[gene_name]:
+ out.append(self.proteins[g])
+ return out
+
+ def get_drug_by_drugbank(self, drugbank_id):
+ return self.drugs[drugbank_id]
+
+ def get_disorder_by_mondo(self, mondo_id):
+ return self.disorders[mondo_id]
\ No newline at end of file
diff --git a/drugstone/models.py b/drugstone/models.py
index 9299122..4f160db 100755
--- a/drugstone/models.py
+++ b/drugstone/models.py
@@ -5,13 +5,6 @@ from django.db import models
# Main biological and medical entities
-class Tissue(models.Model):
- name = models.CharField(max_length=128, default='', unique=True)
-
- def __str__(self):
- return self.name
-
-
class PPIDataset(models.Model):
name = models.CharField(max_length=128, default='', unique=False)
link = models.CharField(max_length=128, default='', unique=False)
@@ -60,18 +53,9 @@ class DrDiDataset(models.Model):
unique_together = ('name', 'version')
-class ExpressionLevel(models.Model):
- tissue = models.ForeignKey('Tissue', on_delete=models.CASCADE)
- protein = models.ForeignKey('Protein', on_delete=models.CASCADE)
- expression_level = models.FloatField()
-
- class Meta:
- unique_together = ('tissue', 'protein')
-
-
-# class EnsemblGene(models.Model):
-# name = models.CharField(max_length=15, unique=True) # starts with ENSG...
-# protein = models.ForeignKey('Protein', on_delete=models.CASCADE, related_name='ensg')
+class EnsemblGene(models.Model):
+ name = models.CharField(max_length=15) # starts with ENSG...
+ protein = models.ForeignKey('Protein', on_delete=models.CASCADE, related_name='ensg')
class Protein(models.Model):
@@ -84,6 +68,7 @@ class Protein(models.Model):
entrez = models.CharField(max_length=15, default='')
drugs = models.ManyToManyField('Drug', through='ProteinDrugInteraction',
related_name='interacting_drugs')
+ ensembl = models.CharField(max_length=15, default='')
tissue_expression = models.ManyToManyField('Tissue', through='ExpressionLevel',
related_name='interacting_drugs')
@@ -106,6 +91,25 @@ class Protein(models.Model):
self.entrez = other.entrez
+class ExpressionLevel(models.Model):
+ tissue = models.ForeignKey('Tissue', on_delete=models.CASCADE)
+ protein = models.ForeignKey('Protein', on_delete=models.CASCADE)
+ expression_level = models.FloatField()
+
+ class Meta:
+ unique_together = ('tissue', 'protein')
+
+ def __hash__(self):
+ return hash(f'{self.tissue_id}_{self.protein_id}')
+
+
+class Tissue(models.Model):
+ name = models.CharField(max_length=128, default='', unique=True)
+
+ def __str__(self):
+ return self.name
+
+
class Disorder(models.Model):
mondo_id = models.CharField(max_length=7)
label = models.CharField(max_length=256, default='') # symbol
@@ -143,7 +147,7 @@ class Drug(models.Model):
return self.drug_id
def __eq__(self, other):
- return self.drug_id == other.uniprot_code and self.name == other.name and self.status == other.status
+ return self.drug_id == other.drug_id and self.name == other.name and self.status == other.status
def __ne__(self, other):
return not self.__eq__(other)
@@ -168,6 +172,15 @@ class ProteinDisorderAssociation(models.Model):
def __str__(self):
return f'{self.pdis_dataset}-{self.protein}-{self.disorder}'
+ def __eq__(self, other):
+ return self.pdis_dataset_id == other.pdis_dataset_id and self.protein_id == other.protein_id and self.disorder_id == other.disorder_id
+
+ def __ne__(self, other):
+ return not self.__eq__(other)
+
+ def __hash__(self):
+ return hash((self.pdis_dataset_id, self.protein_id, self.disorder_id))
+
class DrugDisorderIndication(models.Model):
drdi_dataset = models.ForeignKey(
@@ -181,6 +194,15 @@ class DrugDisorderIndication(models.Model):
def __str__(self):
return f'{self.drdi_dataset}-{self.drug}-{self.disorder}'
+ def __eq__(self, other):
+ return self.drdi_dataset_id == other.drdi_dataset_id and self.drug_id == other.drug_id and self.disorder_id == other.disorder_id
+
+ def __ne__(self, other):
+ return not self.__eq__(other)
+
+ def __hash__(self):
+ return hash((self.drdi_dataset_id, self.drug_id, self.disorder_id))
+
class ProteinProteinInteraction(models.Model):
ppi_dataset = models.ForeignKey(
@@ -210,10 +232,19 @@ class ProteinProteinInteraction(models.Model):
def __str__(self):
return f'{self.ppi_dataset}-{self.from_protein}-{self.to_protein}'
+ def __eq__(self, other):
+ return self.ppi_dataset_id == other.ppi_dataset_id and self.from_protein_id == other.from_protein_id and self.to_protein_id == other.to_protein_id
+
+ def __ne__(self, other):
+ return not self.__eq__(other)
+
+ def __hash__(self):
+ return hash((self.ppi_dataset_id, self.from_protein_id, self.to_protein_id))
+
class ProteinDrugInteraction(models.Model):
pdi_dataset = models.ForeignKey(
- 'PDIDataset', null=True, on_delete=models.CASCADE, related_name='pdi_dataset_relation')
+ PDIDataset, null=True, on_delete=models.CASCADE, related_name='pdi_dataset_relation')
protein = models.ForeignKey('Protein', on_delete=models.CASCADE)
drug = models.ForeignKey('Drug', on_delete=models.CASCADE)
@@ -223,6 +254,15 @@ class ProteinDrugInteraction(models.Model):
def __str__(self):
return f'{self.pdi_dataset}-{self.protein}-{self.drug}'
+ def __eq__(self, other):
+ return self.pdi_dataset_id == other.pdi_dataset_id and self.protein_id == other.protein_id and self.drug_id == other.drug_id
+
+ def __ne__(self, other):
+ return not self.__eq__(other)
+
+ def __hash__(self):
+ return hash((self.pdi_dataset_id, self.protein_id, self.drug_id))
+
class Task(models.Model):
token = models.CharField(max_length=32, unique=True)
diff --git a/docker-entrypoint.sh b/scripts/docker-entrypoint.sh
similarity index 63%
rename from docker-entrypoint.sh
rename to scripts/docker-entrypoint.sh
index 9686a3c..6dc4cd8 100755
--- a/docker-entrypoint.sh
+++ b/scripts/docker-entrypoint.sh
@@ -3,13 +3,17 @@
file="docker-entrypoint.lock"
# exit if entrypoint.lock exists to prevent new import of data every time docker is restarted
-python3 manage.py makemigrations drugstone
-python3 manage.py migrate
+
if ! test -f "$file"; then
+# sh scripts/import-data.sh
+ python3 manage.py makemigrations drugstone
+ python3 manage.py migrate
python3 manage.py createfixtures
python3 manage.py cleanuptasks
- sh scripts/import-data.sh
+ python3 manage.py populate_db -u --all
+ python3 manage.py make_graphs
+
touch $file
fi
diff --git a/scripts/import-data.sh b/scripts/import-data.sh
index de1c938..ea358a3 100755
--- a/scripts/import-data.sh
+++ b/scripts/import-data.sh
@@ -1,11 +1,15 @@
#!/bin/bash
-python3 manage.py populate_db --delete_model PPI,PDI,Drug,Protein,Tissue,Disorder,PDiAssociations
+# python3 manage.py populate_db --delete_model PPI,PDI,PDi,DrDi,Drug,Protein,Tissue,Expression,Disorder
+#python3 manage.py populate_db --all -u
+#python3 manage.py populate_db -p -u
+#python3 manage.py populate_db --clear --all
+#python3 manage.py populate_db --delete_model Disorder
+#python3 manage.py populate_db -u --all
+#python3 manage.py populate_db --delete_model PDI --data_dir . -pdr
+#python3 manage.py populate_db --data_dir . -exp gene_tissue_expression.gct
-python3 manage.py populate_db --data_dir . -p
-python3 manage.py populate_db --data_dir . -exp gene_tissue_expression.gct
-
-python3 manage.py populate_db --data_dir . -dr
+#python3 manage.py populate_db -dr
#python3 manage.py populate_db --data_dir . -pdr drug-protein-interaction.txt
-python3 manage.py populate_db -di
+#python3 manage.py populate_db -di
#python3 manage.py populate_db --data_dir . -pdi "" -ddi ""
#python3 manage.py populate_db -pp protein_protein_interaction_file.txt
\ No newline at end of file
diff --git a/tasks/betweenness_centrality_test.py b/tasks/betweenness_centrality_test.py
index 4edb602..b1a70f8 100755
--- a/tasks/betweenness_centrality_test.py
+++ b/tasks/betweenness_centrality_test.py
@@ -21,7 +21,7 @@ def betweenness_centrality_test(algorithm, parameters):
print()
print(results.get('node_attributes'))
- task_hook = TaskHook(parameters, '../data_drugstone/', set_progress, set_result)
+ task_hook = TaskHook(parameters, '../data/', set_progress, set_result)
algorithm(task_hook)
class Range(object):
diff --git a/tasks/closeness_centrality_test.py b/tasks/closeness_centrality_test.py
index 9275dd4..095b0b3 100755
--- a/tasks/closeness_centrality_test.py
+++ b/tasks/closeness_centrality_test.py
@@ -20,7 +20,7 @@ def closeness_centrality_test(algorithm, parameters):
print()
print(results.get('node_attributes'))
- task_hook = TaskHook(parameters, '../data_drugstone/', set_progress, set_result)
+ task_hook = TaskHook(parameters, '../data/', set_progress, set_result)
algorithm(task_hook)
class Range(object):
diff --git a/tasks/degree_centrality_test.py b/tasks/degree_centrality_test.py
index b082502..10596c5 100755
--- a/tasks/degree_centrality_test.py
+++ b/tasks/degree_centrality_test.py
@@ -20,7 +20,7 @@ def degree_centrality_test(algorithm, parameters):
print()
print(results.get('node_attributes'))
- task_hook = TaskHook(parameters, '../data_drugstone/', set_progress, set_result)
+ task_hook = TaskHook(parameters, '../data/', set_progress, set_result)
algorithm(task_hook)
class Range(object):
diff --git a/tasks/keypathwayminer_test.py b/tasks/keypathwayminer_test.py
index d08fdb7..f3e1383 100755
--- a/tasks/keypathwayminer_test.py
+++ b/tasks/keypathwayminer_test.py
@@ -19,7 +19,7 @@ def task_test(algorithm):
print(f' Edge #{j + 1}: {edge["from"]} -> {edge["to"]}')
print()
- task_hook = TaskHook({'k': 1, 'seeds': ['Q9BS26', 'O00124', 'P33527']}, '../data_drugstone/', set_progress, set_result)
+ task_hook = TaskHook({'k': 1, 'seeds': ['Q9BS26', 'O00124', 'P33527']}, '../data/', set_progress, set_result)
algorithm(task_hook)
diff --git a/tasks/multi_steiner_test.py b/tasks/multi_steiner_test.py
index 6fd8734..dbf5c7e 100755
--- a/tasks/multi_steiner_test.py
+++ b/tasks/multi_steiner_test.py
@@ -20,7 +20,7 @@ def multi_steiner_test(algorithm, parameters):
print()
print(results.get('node_attributes'))
- task_hook = TaskHook(parameters, '../data_drugstone/', set_progress, set_result)
+ task_hook = TaskHook(parameters, '../data/', set_progress, set_result)
algorithm(task_hook)
if __name__ == '__main__':
diff --git a/tasks/network_proximity_test.py b/tasks/network_proximity_test.py
index 1bbb179..9512b0a 100755
--- a/tasks/network_proximity_test.py
+++ b/tasks/network_proximity_test.py
@@ -21,7 +21,7 @@ def network_proximity_test(algorithm, parameters):
print()
print(results.get('node_attributes'))
- task_hook = TaskHook(parameters, '../data_drugstone/', set_progress, set_result)
+ task_hook = TaskHook(parameters, '../data/', set_progress, set_result)
algorithm(task_hook)
diff --git a/tasks/task_hook.py b/tasks/task_hook.py
index 9b8a669..42b860a 100755
--- a/tasks/task_hook.py
+++ b/tasks/task_hook.py
@@ -29,7 +29,7 @@ class TaskHook:
"""
Returns the data directory including trailing slash.
- :return: Data directory (e.g. '/app/data_drugstone/')
+ :return: Data directory (e.g. '/app/data/')
"""
return self.__data_directory
diff --git a/tasks/task_test.py b/tasks/task_test.py
index 206f9e8..7cb9db5 100755
--- a/tasks/task_test.py
+++ b/tasks/task_test.py
@@ -19,7 +19,7 @@ def task_test(algorithm):
print(f' Edge #{j + 1}: {edge["from"]} -> {edge["to"]}')
print()
- task_hook = TaskHook({'seeds': ['Q9BS26', 'O00124', 'P33527']}, '../data_drugstone/', set_progress, set_result)
+ task_hook = TaskHook({'seeds': ['Q9BS26', 'O00124', 'P33527']}, '../data/', set_progress, set_result)
algorithm(task_hook)
diff --git a/tasks/trust_rank_test.py b/tasks/trust_rank_test.py
index 62d2b91..bc9640d 100755
--- a/tasks/trust_rank_test.py
+++ b/tasks/trust_rank_test.py
@@ -20,7 +20,7 @@ def trust_rank_test(algorithm, parameters):
print()
print(results.get('node_attributes'))
- task_hook = TaskHook(parameters, '../data_drugstone/', set_progress, set_result)
+ task_hook = TaskHook(parameters, '../data/', set_progress, set_result)
algorithm(task_hook)
--
GitLab