Personal tools

iRefIndex Release Process

From irefindex
Revision as of 04:44, 19 August 2011 by PaulBoddie (Talk | contribs)

Jump to: navigation, search

The release process consists of a number of activities.


Building the Database

See iRefIndex Build Process for details.

Generating Output Files and Statistics

See iRefIndex Output and Statistics for details.

Testing the Output

Testing includes the following activities:

  • Checking for errors and log messages
  • Web services
  • Spot checks
  • Counting XML tags
  • Validation
    • PSI-MI MITAB files are validated using scripts available at [1]
    • PSI-MI XML files are validated using the validator at [2]
  • Parsing the final output files - see iRefIndex MITAB Parser for details.

See iRefIndex Testing 7.0 for more information.

Posting the Output Files

The output files themselves are made available on the iRefIndex FTP site as described on the iRefIndex information page. This involves making a directory in the following local location:


For example, for the iRefIndex 3.0 release, the following directory hierarchy is created:

mkdir -p /biotek/proteas/ftp/irefindex/data/archive/release_3.0/psimi_tab

The data files are then copied into this new directory. For example:

cp /home/irefindex/output/*.zip /biotek/proteas/ftp/irefindex/data/archive/release_3.0/psimi_tab/

A README file should be placed in the directory containing a link to the format description. For example:

Please see

The symbolic link indicating the location of the current release needs to be updated. For iRefIndex 3.0 this is done as follows:

cd /biotek/proteas/ftp/irefindex/data
rm current
ln -s archive/release_3.0 current

Finally, the release file must be updated to indicate the current release number. For iRefIndex 3.0 this contains the following text:


Optional/Experimental Output Files

Additional PSI-MI-XML files may be made available within the release directory within a subdirectory created as follows (for iRefIndex 3.0):

mkdir -p /biotek/proteas/ftp/irefindex/data/archive/release_3.0/psimi25xml

(Instructions on copying these files will be added in future.)

A README file should be placed in the directory with the following form:

Please see

(This directory contains a sample of iRefIndex data in PSI-MI 2.5 XML format for testing purposes only)


Filter feedback files for level 6 mapping errors and send to databases.

The feedback is generated using BioPSI_Suplimenter as follows:

java -cp build/classes:lib/mysql-connector-java-5.1.6-bin.jar <host> <database> <username> <password> <logfile> <directory> <release>

For example:

java -cp build/classes:lib/mysql-connector-java-5.1.6-bin.jar host irefindex irefindex ... /home/irefindex/logs/feedback.log /home/irefindex/output/ 9.0

This activity must be completed before iRefWeb data is prepared due to some dependencies on some of this activity's products.

Updating the Wiki

The iRefIndex page will need updating with version-specific references such as the following:

  • README_iRefIndex_MITAB_x.y
  • Sources_iRefIndex_x.y
  • README_iRefIndex_Feedback_x.y
  • Statistics_iRefIndex_x.y

New documents will need creating for some of these references, and general redirect pages should be updated to point to the latest versions of the above pages.

For new data sources, the text will need editing and a citation reference should be added for the new source.


Data for iRefScape must be prepared.


Data for iRefWeb must be prepared as described on the iRefIndex Data Preparation for iRefWeb page.

iRefWeb LinkOut Data

The iRefWeb LinkOut Generator needs to be run and the generated data submitted to the LinkOut FTP site once iRefWeb has been updated.

All iRefIndex Pages

Follow this link for a listing of all iRefIndex related pages (archived and current).