Chapter 10 - Setting Up the SHRINE Ontology

This chapter goes through the steps that are involved in loading the 'demo' SHRINE ontology and setting up the corresponding adapter mappings file. In network specific setups, you will be loading a network specific ontology and will be mapping terms specific to that ontology.

The first thing you will need to do is to create the shrine_ont database user within i2b2 and a schema with the same name:

$ psql -U postgres i2b2  

create user shrine_ont with password 'demouser';
create schema authorization shrine_ont;

grant all privileges on all tables in schema shrine_ont to shrine_ont;
grant all privileges on all sequences in schema shrine_ont to shrine_ont;
grant all privileges on all functions in schema shrine_ont to shrine_ont;

Next, you will need to create the tables that will contain the SHRINE ontology. Please refer to this file for the script: ontology_create_tables.sql

You can run this script by running:

$ psql -U postgres "dbname=i2b2 options=--search_path=shrine_ont" -a -f ontology_create_tables.sql

Next, you will need to download the demo SHRINE SCHEMES file to load into the shrine_ont schema. The file is hosted on our SVN repository here:

https://open.med.harvard.edu/svn/shrine-ontology/SHRINE_Demo_Downloads/trunk/SCHEMES.sql

Once you have downloaded the script, you can run this script by running:

$ psql -U postgres "dbname=i2b2 options=--search_path=shrine_ont" -a -f SCHEMES.sql

Next, you will need to load the demo SHRINE ontology into the newly created tables. This will populate the ontology tree within the SHRINE webclient. You can download the demo ontology file here:

https://open.med.harvard.edu/svn/shrine-ontology/SHRINE_Demo_Downloads/trunk/ShrineDemo.sql

Once you have downloaded the script, you can run this script by running:

$ psql -U postgres "dbname=i2b2 options=--search_path=shrine_ont" -a -f ShrineDemo.sql

Next, you will need to point i2b2 to look at the newly created schemas in its database for the SHRINE ontology. Please refer to this file for the script: db_lookups.sql

You can run this script by running:

$ psql -U postgres "dbname=i2b2 options=--search_path=i2b2hive" -a -f db_lookups.sql

Lastly, you will need to configure the table access within the shrine_ont schema:

$ psql -U postgres i2b2
 
i2b2=# set search_path = shrine_ont;
 
i2b2=# INSERT into shrine_ont.TABLE_ACCESS ( C_TABLE_CD, C_TABLE_NAME, C_PROTECTED_ACCESS, C_HLEVEL, C_NAME, C_FULLNAME, C_SYNONYM_CD, C_VISUALATTRIBUTES, C_TOOLTIP, C_FACTTABLECOLUMN, C_DIMTABLENAME, C_COLUMNNAME, C_COLUMNDATATYPE, C_DIMCODE, C_OPERATOR) values ( 'SHRINE', 'SHRINE', 'N', 0, 'SHRINE Ontology', '\SHRINE\', 'N', 'CA', 'SHRINE Ontology', 'concept_cd', 'concept_dimension', 'concept_path', 'T', '\SHRINE\', 'LIKE');