add omop id to active moiety entity graph

Author: Richard Boyce

linkedSPLs/LinkedSPLs-activeMoiety/README

@@ -6,17 +6,19 @@ This folder will hold a graph that maps the active moiety components of an SPL t
 
 Inputs are original mappings listed as belows:
 
-(1) UNII to RxNORM
+(1) PT to UNII
 
-(2) UNII to ChEBI
+(2) UNII to RxNORM
 
-(3) RxNORM to Drugbank
+(3) PT to Drugbank
 
-(5) RxNORM to NDFRT (EPC)
+(4) PT to ChEBI
 
-(6) RxNORM to OMOP
+(5) RxNORM to OMOP
 
-(6) PreferredTerm to drug class
+(6) RxNORM to DrOn
+
+(7) RxNORM to NDFRT (EPC)
 
 Outputs is a RDF/XML graph that represents all active moiety with linked resouces.
 
@@ -28,12 +30,24 @@ Procedures to get active moieties RDF graph
 <STEP 1>: prepare mappings in folder mappings/
 
 ----------------------------------------------
-mappings/dron#to#chebi#and#rxnorm.txt
+mappings of PT to UNII (FDA)
+----------------------------------------------
+
+FROM "../../LinkedSPLs-update/data/FDA/FDAPreferredSubstanceToUNII.txt"
+
+----------------------------------------------
+mappings of UNII to RXCUI (UMLS)
+----------------------------------------------
+
+FROM "../../LinkedSPLs-update/data/UMLS/UNIIs-Rxcuis-from-UMLS.txt"
+
+----------------------------------------------
+mappings/dron-to-chebi-and-rxnorm.txt
 ----------------------------------------------
-FROM "../../LinkedSPLs#update/mappings/DrOn#to#RxNorm/dron#to#chebi#and#rxnorm.txt"
+FROM "../../LinkedSPLs-update/mappings/DrOn-to-RxNorm/dron-to-chebi-and-rxnorm.txt"
 
 ----------------------------------------------
-mappings/imeds_drugids_to_rxcuis.csv
+mappings/active-ingredient-omopid-rxcui-09042015.dsv
 ----------------------------------------------
 
 query OMOP CDM V5 (GeriOMOP) by SQL query below:
@@ -47,24 +61,19 @@ cpt.CONCEPT_CLASS = 'Ingredient';
 right click result table and export to csv ('|' delimited) 
 
 ----------------------------------------------
-mappings/UNIIToChEBI#<DATE>.txt
-----------------------------------------------
-FROM "../../LinkedSPLs#update/mappings/PT#UNII#ChEBI#mapping/UNIIToChEBI#<DATE>.txt"
-
-----------------------------------------------
-mappings/pt_drugbank#<DATE>.txt
+mappings/UNIIToChEBI-<DATE>.txt
 ----------------------------------------------
-FROM "../../LinkedSPLs#update/mappings/ChEBI#DrugBank#bio2rdf#mapping/fda#substance#preferred#name#to#drugbank#<DATE>.txt"
+FROM "../../LinkedSPLs-update/mappings/PT-UNII-ChEBI-mapping/UNIIToChEBI-<DATE>.txt"
 
 ----------------------------------------------
-mappings/PreferredTerm#UNII#Rxcui#mapping.txt
+mappings/pt_drugbank-<DATE>.txt
 ----------------------------------------------
-FROM "../../LinkedSPLs#update/mappings/RxNORM#mapping/PreferredTerm#UNII#Rxcui#mapping.txt"
+FROM "../../LinkedSPLs-update/mappings/ChEBI-DrugBank-bio2rdf-mapping/fda-substance-preferred-name-to-drugbank-<DATE>.txt"
 
 ----------------------------------------------
 mappings/EPC_extraction_most_recent_<DATE>.txt"
 ----------------------------------------------
-FROM "../../LinkedSPLs#update/mappings/pharmacologic_class_indexing/EPC_extraction_most_recent_<DATE>.txt"
+FROM "../../LinkedSPLs-update/mappings/pharmacologic_class_indexing/EPC_extraction_most_recent_<DATE>.txt"
 
 
 <STEP 2>: run python script to merge those mappings together
@@ -94,7 +103,7 @@ from "http://biopython.org/wiki/Download"
 (2) rdflib 
 
 (3) rdflib#jsonld
-from "https://github.com/RDFLib/rdflib#jsonld"
+from "https://github.com/RDFLib/rdflib-jsonld"
 
 (4) pandas
 

linkedSPLs/LinkedSPLs-activeMoiety/mergeToActiveMoiety.py

@@ -11,71 +11,76 @@
 import pandas as pd
 from StringIO import StringIO
 import numpy as np
-#from numpy import nan
 
+## Define namespaces
 CHEBI_BASE_URI = "http://purl.obolibrary.org/obo/"
 RXNORM_BASE_URI = "http://purl.bioontology.org/ontology/RXNORM/"
-## Define data inputs
 
-#UNIIS = "../linkedSPLs-update/data/UMLS/UNIIs-from-UMLS.txt"
-#PT_UNII = "mappings/FDAPreferredTermToUNII.tsv"
-#PT_RXCUI = "mappings/fda-active-moiety-string-name-rxnorm-mapping.csv"
+## Define data inputs
+PT_UNII = "../LinkedSPLs-update/data/FDA/FDAPreferredSubstanceToUNII.txt"
+UNII_RXCUI = "../LinkedSPLs-update/data/UMLS/UNIIs-Rxcuis-from-UMLS.txt"
 
 PT_CHEBI = "mappings/UNIIToChEBI-06102015.txt"
 PT_DRUGBANK = "mappings/fda-substance-preferred-name-to-drugbank-06102015.txt"
-UNII_PT_RXCUI = "mappings/PreferredTerm-UNII-Rxcui-mapping.txt"
+
 UNII_NUI_PREFERRED_NAME_ROLE = "mappings/EPC_extraction_most_recent_06102015.txt"
 DRON_CHEBI_RXCUI = "mappings/cleaned-dron-chebi-rxcui-ingredient-06222015.txt"
-#OMOP_RXCUI = "mappings/imeds_drugids_to_rxcuis.csv"
+OMOP_RXCUI = "mappings/active-ingredient-omopid-rxcui-09042015.dsv"
 
+## Get UNII - PT - RXCUI
+unii_pt_cols = ['unii','pt']
+unii_pt_DF = pd.read_csv(PT_UNII, sep='\t', names=unii_pt_cols)
 
-## read mappings of pt, unii and rxcui
-
-unii_pt_rxcui_cols = ['unii','pt','rxcui']
-unii_pt_rxcui_DF = pd.read_csv(UNII_PT_RXCUI, sep='\t', names=unii_pt_rxcui_cols, skiprows=[0])
+rxcui_unii_cols = ['rxcui','unii']
+rxcui_unii_DF = pd.read_csv(UNII_RXCUI, sep='|', names=rxcui_unii_cols)
+rxcui_unii_DF['rxcui'] = rxcui_unii_DF['rxcui'].astype('str')
 
+unii_pt_rxcui_DF = unii_pt_DF.merge(rxcui_unii_DF, on=['unii'], how='left')
+print unii_pt_rxcui_DF.info()
 
 ## read mappings of pt and drugbank uri
 
 pt_drugbank_cols = ['pt','db_uri1','db_uri2']
 pt_drugbank_DF = pd.read_csv(PT_DRUGBANK, sep='\t', names=pt_drugbank_cols)
 
-
 ## read mappings of pt and chebi
 
 pt_chebi_cols = ['pt','chebi']
 pt_chebi_DF = pd.read_csv(PT_CHEBI, sep='\t', names=pt_chebi_cols)
-#print pt_chebi_DF.info()
 
 ## read mappings of dron and rxcui
 dron_chebi_rxcui_cols = ['dron','chebi','rxcui']
 dron_chebi_rxcui_DF = pd.read_csv(DRON_CHEBI_RXCUI, sep='|', names=dron_chebi_rxcui_cols, usecols=[0,2])
+dron_chebi_rxcui_DF['rxcui'] = dron_chebi_rxcui_DF['rxcui'].astype('str')
 
 ## read mappings of unii, nui and preferredNameAndRole
 unii_nui_namerole_cols = ['setid', 'unii','nui','nameAndRole']
 unii_nui_namerole_DF = pd.read_csv(UNII_NUI_PREFERRED_NAME_ROLE, sep='\t', names=unii_nui_namerole_cols)[['unii','nui','nameAndRole']]
-#print unii_nui_namerole_DF.info()
 
 
+## read mappings of omopid and rxcui
+omop_rxcui_cols = ['omopid','rxcui']
+omop_rxcui_DF = pd.read_csv(OMOP_RXCUI, sep='|', names=omop_rxcui_cols)
+omop_rxcui_DF['rxcui'] = omop_rxcui_DF['rxcui'].astype('str')
+
 ## merge pt, unii, rxcui and drugbank uri
 unii_pt_rxcui_db_DF = unii_pt_rxcui_DF.merge(pt_drugbank_DF, on=['pt'], how='left')
-#print unii_pt_rxcui_db_DF.info()
-unii_pt_rxcui_db_DF.to_csv('PT-RXCUI-UNII-DB.csv', sep='\t', index=False)
 
+unii_pt_rxcui_db_DF.to_csv('PT-RXCUI-UNII-DB.csv', sep='\t', index=False)
 
 ## merge chebi 
-
 merged_chebi_DF = unii_pt_rxcui_db_DF.merge(pt_chebi_DF, on=['pt'], how='left')
-#print merged_chebi_DF.info()
 
 ## merge dron id
 merged_dron_DF = merged_chebi_DF.merge(dron_chebi_rxcui_DF, on=['rxcui'], how = 'left')
 
+## merge omop id
+merged_omop_DF = merged_dron_DF.merge(omop_rxcui_DF, on=['rxcui'], how = 'left')
+
 ## merge <nui> and <preferred name and role>
-merged_epc_DF = merged_dron_DF.merge(unii_nui_namerole_DF, on=['unii'], how='left')
+merged_epc_DF = merged_omop_DF.merge(unii_nui_namerole_DF, on=['unii'], how='left')
 
 print merged_epc_DF.info()
 
-
 merged_epc_DF.to_csv('mergedActiveMoiety.csv', sep='\t', index=False)