wiki:Using I2B2

Our HERON data repository is based on I2B2:

i2b2 (Informatics for Integrating Biology and the Bedside) is an NIH-funded National Center for Biomedical Computing based at Partners HealthCare System. The i2b2 Center is developing a scalable informatics framework that will enable clinical researchers to use existing clinical data for discovery research and, when combined with IRB-approved genomic data, facilitate the design of targeted therapies for individual patients with diseases having genetic origins

See I2B2Community regarding mailing lists, conferences, upstream collaboration etc.


Corresponding presentation:

i2b2 at KUMC

In addition to the main installation, we have other dev/test installs. See GroupOnly/Operations for details.

i2b2 installation

In addition to the AUG e-mail repository, there is also i2b2-install-help Google Group which may be a helpful reference.

source:i2b2-backend-deploy automates the installation of the i2b2 cells that we use:

  1. PM Cell
  2. Ontology Cell
  3. CRC aka Data Repository Cell
  4. Workplace Cell

The HERON i2b2 hive (back end) deployment generated documentation should help some, though it's not a complete substitute for reading the code.

See #756, #1087 for status.

See HeronLoad#heron_init for the way we created our schemas and such; it runs the SQL scripts from i2b2 to create datamarts etc.

For database password management, see also the configure_i2b2_datasources task under HeronLoad#release .

Apache, JBoss, and i2b2

The parts look something like this:

GraphViz image


attachment:Doc_for_i2b2_1.4_Install.doc is Arvinder's documentation on the original installation at KUMC in early 2010, along with attachment:Design_document_i2b2.doc (system architecture). It was a significant learning process regarding many dependencies:

  1. Java JDK: JDK RPM Installation for Linux (64-bit) from Oracle (formerly Sun)
  2. JBoss 4.2.2 GA: i2b2 1.4 core docs
  3. Apache Ant: directory layout from Installing Ant
  4. Apache Axis2 1.1: i2b2 core docs and Apache Axis2 Installation Guide
  5. Oracle 10g: source(s) not yet tracked down
  6. Apache Web Server: How to install Apache on SuSE Linux by richard laffers
  7. php: How to install PHP on SuSE Linux by richard laffers
    • note --with-curl config option

attachment:Creating_a_Project_in_i2b2.doc has been partially automated; see create_i2b2_project() in source:i2b2-backend-deploy/

Starting i2b2 application

Now that #134 and #394 have been addressed...

The normal /etc/init.d/jboss start mechanism starts jboss, and hence the i2b2 hive, at boot time. To stop, use the conventional /etc/init.d/jboss stop. The ssh tunnels necessary to connect i2b2 to the database start automatically at boot as well.

How do i2b2 Projects relate to the Enterprise Repository?

In How to i2b2 (from AMIA March 2010 Summit among i2b2 events):

  • on slide 8, we see research cohorts in "project silos"
  • in slide 20 "Upload data through i2b2 Workbench" @@@
  • slide 9 has a nice figure:

slide 9 "Project data can be added back to the ESD" from "How to i2b2"

There is some discussion of this in the the 2010 i2b2 paper.

In members@…:

"Murphy, Shawn N." <SNMURPHY@…> 7/8/2010 5:25 PM >>>
Hi Keith,

Indeed, the current design is for each project to have its own data schema.
This is so we can cleanly manage user permissions and data ownership when
derived data is created for a project. This can be facilitated with a package
of sql scripts that can create the new schema and move the data from an existing
schema for specific patient sets and data types. They are under the source_code
section of the software page , in a package named


i2b2 and R

See HERON Bow Creek Release brings Cancer Survival Analysis, R integration Feb 2012

based on:

Last modified 3 years ago Last modified on 01/21/14 11:18:40

Attachments (4)