Merge branch 'oncokb' into string

Author: franbeb

DEPLOYING.md

@@ -37,7 +37,7 @@ BioAPI uses three genomic databases for its operation. These databases must be l
 
 To import all databases in MongoDB:
  
-1. Download the "bioapi_db.gz" from **[here](https://drive.google.com/file/d/1lI3A98N-GhnffkSOWjB_gx_ieq3pEjFP/view?usp=sharing)**
+1. Download the "bioapi_db.gz" from **[here](https://drive.google.com/file/d/1oBdhC-XoJn-VNEIEpfMWB2Gna--WZ1Wa/view?usp=sharing)**
 2. Shutdown all the services running `docker compose down`
 3. Edit the `docker-compose.yml` file to include the downloaded file inside the container:
     ```yml

bio-api/bioapi.py

@@ -3,6 +3,7 @@
 import json
 import gzip
 import logging
+from db import get_mongo_connection
 from concurrent.futures import ThreadPoolExecutor
 import configparser
 import urllib.parse
@@ -41,16 +42,17 @@
 # Common response header
 headers = {"Content-Type": "application/json"}
 
-from db import get_mongo_connection
-mydb = get_mongo_connection(IS_DEBUG,Config)
 
+mydb = get_mongo_connection(IS_DEBUG, Config)
+
+def get_potential_gene_symbols(query_string: str, limit_elements: int = 50) -> List[str]:
 
-def get_potential_gene_symbols(query_string, limit_elements):
     """
-    TODO: document and add types
-    :param query_string:
-    :param limit_elements:
-    :return:
+    Takes a string of any length and returns a list of genes that contain that search criteria.
+
+    :param query_string: The query string
+    :param limit_elements: The number of elements to return
+    :return: A list of genes that contain that search criteria
     """
     er = re.compile("^" + re.escape(query_string), re.IGNORECASE)
     collection_hgnc = mydb["hgnc"]  # HGNC collection
@@ -83,11 +85,12 @@ def get_potential_gene_symbols(query_string, limit_elements):
     return res
 
 
-def search_gene_group(gen):  # AGREGAR LO QUE PASA SI NO PERTENECE A NINGUN gene_group_id (EJ gen:AADACP1)
+def search_gene_group(gen: str) -> Dict[str, Any]:
     """
-    TODO: document
-    :param gen:
-    :return:
+    Get the gene and locus group from HGNC DB for a specific approved gene symbol
+
+    :param gen: an approved gene symbol
+    :return: a dictionary with the gene and locus group information
     """
     results = {'locus_group': None, 'locus_type': None, 'gene_group': None, 'gene_group_id': None}
     collection_hgnc = mydb["hgnc"]  # HGNC collection
@@ -107,17 +110,18 @@ def search_gene_group(gen):  # AGREGAR LO QUE PASA SI NO PERTENECE A NINGUN gene
                 results['gene_group'] = document['gene_group']
                 results['gene_group_id'] = document['gene_group_id']
             else:
-                results['gene_group'] = [document['gene_group']] # type: ignore
-                results['gene_group_id'] = [document['gene_group_id']] # type: ignore
+                results['gene_group'] = [document['gene_group']]  # type: ignore
+                results['gene_group_id'] = [document['gene_group_id']]  # type: ignore
 
     return results
 
 
-def search_genes_in_same_group(group_id: int):
+def search_genes_in_same_group(group_id: int) -> List[str]:
     """
-    TODO: document
-    :param group_id:
-    :return:
+    From a gene group id in HGNC DB, get a list of all genes in the same group.
+
+    :param group_id: a valid gene group id from HGNC DB
+    :return: a list of all genes that are in the group
     """
     collection_hgnc = mydb["hgnc"]  # HGNC collection
     query = {'gene_group_id': group_id}
@@ -126,12 +130,13 @@ def search_genes_in_same_group(group_id: int):
     return [doc["symbol"] for doc in docs]
 
 
-def get_genes_of_pathway(pathway_id, pathway_source):
+def get_genes_of_pathway(pathway_id: str, pathway_source: str) -> List[str]:
     """
-    TODO: document and add types
-    :param pathway_id:
-    :param pathway_source:
-    :return:
+    From a gene group id in HGNC DB, get a list of all genes in the same group.
+
+    :param pathway_source: the pathway database
+    :param pathway_id: an pathway id to search in the database
+    :return: a list of all genes in the pathway
     """
     collection_cpdb = mydb["cpdb"]  # CPDB collection
     ps = re.compile("^" + pathway_source + "$", re.IGNORECASE)
@@ -140,11 +145,12 @@ def get_genes_of_pathway(pathway_id, pathway_source):
     return doc["hgnc_symbol_ids"] if doc is not None else []
 
 
-def get_pathways_of_gene(gene):
+def get_pathways_of_gene(gene: str) -> List[str]:
     """
-    TODO: document and add types
-    :param gene:
-    :return:
+    Get all pathways and sources for a given gene symbols.
+
+    :param gene: apprved gene symbol
+    :return: list of pathways
     """
     collection_cpdb = mydb["cpdb"]  # CPDB collection
     query = {'hgnc_symbol_ids': gene}
@@ -153,11 +159,12 @@ def get_pathways_of_gene(gene):
     return [str(doc) for doc in docs]
 
 
-def get_information_of_genes(genes: List[str]) -> Dict:
+def get_information_of_genes(genes: List[str]) -> Dict[str, Dict[str, Any]]:
     """
-    TODO: document
-    :param genes:
-    :return:
+    This function receives a list of gene symbols and returns information about them from different bioinformatics databases.
+
+    :param genes: list of gene symbols
+    :return: dictionary with information about gene symbols (each key is a gene and each value is another dictionary with its information)
     """
     res = {}
     collection_gene_grch37 = mydb["gene_grch37"]
@@ -212,12 +219,13 @@ def get_information_of_genes(genes: List[str]) -> Dict:
     return res
 
 
-def get_expression_from_gtex(tissue: str, genes: List[str]) -> List:
+def get_expression_from_gtex(tissue: str, genes: List[str]) -> List[Dict[str, float]]:
     """
-    Gets all the expressions for a specific tissue and a list of genes
+    Gets all the expressions for a specific tissue and a list of genes.
+
     :param tissue: Tissue to filter
     :param genes: List of genes to filter
-    :return: List of expressions
+    :return: List of expressions values. Each element of the list contains the expression values for each gene
     """
     collection = mydb["gtex_" + tissue]  # Connects to specific tissue's collection
     query = {'gene': {'$in': genes}}
@@ -243,7 +251,7 @@ def terms_related_to_one_gene(gene: str, relation_type: Optional[List[str]] = No
     if relation_type is None:
         relation_type = ["enables", "involved_in", "part_of", "located_in"]
     collection_go_annotations = mydb["go_anotations"]
-    
+
     annotation = list(collection_go_annotations.find({"gene_symbol": gene}))
     related_genes = {}
     if annotation:
@@ -260,6 +268,7 @@ def terms_related_to_one_gene(gene: str, relation_type: Optional[List[str]] = No
     return related_genes
 
 
+
 def is_term_on_db(term_id)-> bool:
     """
     Returns whether a go term ID is in the DB
@@ -361,12 +370,12 @@ def enrich(gene_ids: List, p_value_threshold: int= 0.05, correction_method: str
     return metrics, relations
 
 
-def populate_terms_with_data(term_list:List, ontology_type: Optional[List[str]] = None)-> List:
+def populate_terms_with_data(term_list, ontology_type: Optional[List[str]] = None):
     """
-    Given a go terms list will return a list with all the terms information
-    :param term_list: list of all the wanted terms
-    :param ontology_type: Filters the ontology type of the terms in the response
-    :return: list of all terms with all the information of them that's on the DB
+    TODO: document
+    :param term_list:
+    :param ontology_type:
+    :return:
     """
     if ontology_type is None:
         ontology_type = ["biological_process", "molecular_function", "cellular_component"]
@@ -375,6 +384,7 @@ def populate_terms_with_data(term_list:List, ontology_type: Optional[List[str]]
     return terms
 
 
+
 def strip_term(term: Dict,relations:Optional[List[str]])-> Dict:
     """
     Given a go term and wanted relations will return just a selected amount of attributes needed for representation as a graph
@@ -494,12 +504,12 @@ def cancer_drugs_related_to_gene(gene: str) -> List:
     collection_pharm = mydb["pharmgkb"]
     return list(collection_pharm.find({"genes":gene},{"_id":0}))
 
-# App
 
-def get_data_from_oncokb(genes: List[str]) -> Dict:
+def get_data_from_oncokb(genes: List[str]) -> Dict[str, Dict[str, Any]]:
     """
-    Gets all data associated with a gene list.
-    :param genes: List of genes to filter.
+    Gets all data from OncoKB database associated with a gene list.
+
+    :param genes: List of gene symbols.
     :return: Dict of genes with their associated drugs and information according to OncoKB database
     """
     collection_actionability_gene = mydb["oncokb_biomarker_drug_associations"]
@@ -528,10 +538,10 @@ def get_data_from_oncokb(genes: List[str]) -> Dict:
             res[gen]["oncokb_cancer_gene"].append("Oncogene")
         if doc_c["tumor_suppressor_gene"]:
             res[gen]["oncokb_cancer_gene"].append("Tumor Suppressor Gene")
-        
+
         if len(res[gen]["oncokb_cancer_gene"]) == 0:
             res[gen].pop("oncokb_cancer_gene")
-        
+
         sources = []
         for key in doc_c:
             if doc_c[key] == 1:
@@ -619,7 +629,7 @@ def gene_symbols():
     @flask_app.route("/gene-symbols-finder/", methods=['GET'])
     def gene_symbol_finder():
         """Takes a string of any length and returns a list of genes that contain that search criteria."""
-        query = None  # To prevent MyPy warning
+        query = ""  # To prevent MyPy warning
         if "query" not in request.args:
             abort(400, "'query' parameter is mandatory")
         else:
@@ -642,7 +652,8 @@ def gene_symbol_finder():
     @flask_app.route("/information-of-genes", methods=['POST'])
     def information_of_genes():
         """Receives a list of gene IDs and returns information about them."""
-        body = request.get_json() # type: ignore
+        body = request.get_json()  # type: ignore
+        response = {}
         if "gene_ids" not in body:
             abort(400, "gene_ids is mandatory")
 
@@ -653,7 +664,6 @@ def information_of_genes():
         try:
             response = get_information_of_genes(gene_ids)
         except Exception as e:
-            response = {}  # To prevent mypy warnings
             abort(400, e)
         return make_response(response, 200, headers)
 
@@ -695,7 +705,7 @@ def genes_in_the_same_group(gene_id):
     def pathway_genes(pathway_source, pathway_id):
         if pathway_source.lower() not in PATHWAYS_SOURCES:
             abort(404, f'{pathway_source} is an invalid pathway source')
-        response = {"genes": get_genes_of_pathway(pathway_id, pathway_source)}
+        response = {"genes": get_genes_of_pathway(str(pathway_id), pathway_source)}
         return make_response(response, 200, headers)
 
     @flask_app.route("/pathways-in-common", methods=['POST'])
@@ -720,7 +730,7 @@ def pathways_in_common():
 
     @flask_app.route("/expression-of-genes", methods=['POST'])
     def expression_data_from_gtex():
-        body = request.get_json() # type: ignore
+        body = request.get_json()  # type: ignore
 
         if "gene_ids" not in body:
             abort(400, "gene_ids is mandatory")

config/nginx/conf.d/default.conf

@@ -9,13 +9,18 @@ upstream web {
 
 server {
     client_max_body_size 0;
+
+    # Prevents 504 error with long running requests (3600 seg = 1 hour)
     proxy_read_timeout 3600;
     proxy_connect_timeout 3600;
     proxy_send_timeout 3600;
 
     # Main proxy
     location / {
         proxy_pass http://web/;
+
+        # Prevents 504 error with long running requests
+        uwsgi_read_timeout 3600; # 1 hour
     } 
 
     # Listening port and server name