New data on May 2024

manifest.json

@@ -8,7 +8,7 @@
     "url" : "https://github.com/rjawesome"
   },
   "dumper" : {
-      "data_url" : ["https://www.bindingdb.org/bind/downloads/BindingDB_All_2022m5.tsv.zip"],
+      "data_url" : ["https://www.bindingdb.org/bind/downloads/BindingDB_All_202405_tsv.zip"],
       "uncompress" : true
   },
   "uploader" : {

parser.py

@@ -1,13 +1,21 @@
 import csv
-import json
 import os
 from typing import Dict
 
+
+"""
+Dumped file name
+"""
+DATA_FILE_NAME = 'BindingDB_All_202405.tsv'
+
+
 """
 Fields of the Imported CSV:
 
 - The array BASE_COLS defines the first columns in each row with data mainly pertaining to the compound.
 - A sequence of columns is repeated which creates the data for each protein in the row, which is stored in REPEAT_SUBJECT_COLS.
+
+- Reference: https://www.bindingdb.org/bind/chemsearch/marvin/BindingDB-TSV-Format.pdf
 """
 REPEAT_SUBJECT_COLS = [
     'BindingDB Target Chain Sequence',
@@ -30,7 +38,7 @@
     'Ligand InChI Key',
     'BindingDB MonomerID',
     'BindingDB Ligand Name',
-    'Target Name Assigned by Curator or DataSource',
+    'Target Name',
     'Target Source Organism According to Curator or DataSource',
     'Ki (nM)',
     'IC50 (nM)',
@@ -42,6 +50,7 @@
     'Temp (C)',
     'Curation/DataSource',
     'Article DOI',
+    'BindingDB Entry DOI',
     'PMID',
     'PubChem AID',
     'Patent Number',
@@ -109,7 +118,7 @@
       "uniprot_type": "all",
       "relation": False
     },
-    "Target Name Assigned by Curator or DataSource": {
+    "Target Name": {
       "location": "subject.name",
       "type": "split_colon",
       "uniprot_type": "all",
@@ -181,6 +190,12 @@
       "uniprot_type": "all",
       "relation": True
     },
+    "BindingDB Entry DOI": {
+      "location": "relation.bindingdb_entry_doi",
+      "type": "string",
+      "uniprot_type": "all",
+      "relation": True
+    },
     "PMID": {
       "location": "relation.pmid",
       "type": "string",
@@ -327,13 +342,13 @@
     },
     "UniProt (SwissProt) Secondary ID(s) of Target Chain": {
       "location": "subject.uniprot.secondary_accession",
-      "type": "split_comma",
+      "type": "split_space",
       "uniprot_type": "swissprot",
       "relation": False
     },
     "UniProt (SwissProt) Alternative ID(s) of Target Chain": {
       "location": "subject.uniprot.alternative_accession",
-      "type": "split_comma",
+      "type": "split_space",
       "uniprot_type": "swissprot",
       "relation": False
     },
@@ -357,13 +372,13 @@
     },
     "UniProt (TrEMBL) Secondary ID(s) of Target Chain": {
       "location": "subject.uniprot.secondary_accession",
-      "type": "split_comma",
+      "type": "split_space",
       "uniprot_type": "trembl",
       "relation": False
     },
     "UniProt (TrEMBL) Alternative ID(s) of Target Chain": {
       "location": "subject.uniprot.alternative_accession",
-      "type": "split_comma",
+      "type": "split_space",
       "uniprot_type": "trembl",
       "relation": False
     }
@@ -423,6 +438,8 @@ def process_field(field_name: str, value: str):
     field_type = COLUMN_DATA[field_name]['type']
     if field_type == "int":
         return int(value)
+    if field_type == "split_space":
+        return value.split(' ')
     if field_type == "split_comma":
         return value.split(',')
     if field_type == "split_semicolon":
@@ -446,13 +463,13 @@ def read_csv(file: str, delim: str):
                 continue
             else:
                 base = {'object': {}, 'subject': {}, 'relation': {}}
-                for j in range(37):
+                for j in range(38):
                     if row[j] is not None and row[j] != '' and row[j] != 'NULL':
                         val = process_field(BASE_COLS[j], row[j])
                         set_field(base, BASE_COLS[j], val)
 
-                repeats = int(row[36])  # Number of Protein Chains in Target
-                pos = 37
+                repeats = int(row[37])  # Number of Protein Chains in Target
+                pos = 38
                 for j in range(repeats):
                     info_1 = special_copy(base)
                     info_2 = special_copy(base)
@@ -499,8 +516,7 @@ def merge(main: Dict[str, any], other: Dict[str, any]):
 def load_data(data_folder):
     docs = {}
     row_num = 0
-    for row in read_csv(os.path.join(data_folder, './BindingDB_All.tsv'), '\t'):
-        # print(row['subject']['uniprot'])
+    for row in read_csv(os.path.join(data_folder, './' + DATA_FILE_NAME), '\t'):
         try:
             entry_name = row['subject']['uniprot']['id']
             primary_id = row['subject']['uniprot']['accession']
@@ -531,7 +547,7 @@ def load_data(data_folder):
         yield docs[doc_id]
 
 
-# def main():
+# type: ignore # def main():
 #     from time import time
 
 #     cnt = 0