Skip to content

InterMine 1.4

Compare
Choose a tag to compare
@julie-sullivan julie-sullivan released this 15 Jan 11:38
· 4694 commits to master since this release

We are happy to announce the latest release of InterMine.

We've added some advanced features and as such do NOT recommend you upgrade your InterMine installation unless you are using Tomcat 7 and Postgres 9.x.

This release does not have any model changes but there are some config updates you must do. See the upgrade instructions for details.

Hikari

The biggest change for this InterMine version is the introduction of the Hikari connection pool. Previously InterMine used the default, basic one; you will notice a performance improvement. Be sure to update your maxConnections parameters -- see the upgrade instructions for details.

Even Faster!

In addition to a general increase in performance from Hikari, we made some changes to speed up certain queries, e.g. LOOKUP queries with wildcards, bioseg, range constraints, etc

More Testing Tools

- Continuous Integration - Travis. - [InterMine's](https://travis-ci.org/intermine/intermine/) - runs every time someone commits to the InterMine Git repository - [how to get your own](http://intermine.readthedocs.org/en/latest/get-started/intermine-tests/) - We've added a slew of new tests, including fancy [Selenium](http://www.seleniumhq.org/) tests. See [here](http://intermine.readthedocs.org/en/latest/get-started/intermine-tests/) for the documentation. - We also added several new scripts for setting up the testmodel. - The LOG files record the query being run but in IQL, the InterMine Query Language. You can now run these queries directly in the console in mine/dbmodel

ant run-iql-query -Dquery='some IQL'

Query for empty references or collections

You've always been to query for empty attributes, but now you can search for relationships that don't exist, e.g. "Give me all the genes without a publication." Here is the XML version of that query:

<query name="" model="genomic" view="Gene.secondaryIdentifier" longDescription="" sortOrder="Gene.secondaryIdentifier asc">
  <constraint path="Gene.publications" op="IS NULL"/>
</query>

No More Bioseg!

Previously we wrote our own index type to use for range queries, but now these have finally been introduced in PostgreSQL 9.2. If you are using this version of PostgreSQL or later you do not need to install Bioseg. You still need to install GIST however. Bioseg still works and is performant.

OpenAuth2

You can configure your mine to accept delegated authentication from one or more identity resources which are protected by OAuth2 authentication

Data fixes

- #782 - NCBI updated how they format some identifiers - #742 - Merging issues with human homologues - #829 - BioPAX file name change - #809 - BioGRID: append MGI prefix - #781 - Publications: Process consortiums as authors correctly. - #830 - Throw exception with helpful error message on finding ontology term without a name

Other features / fixes

- Add organism to link to your mine for more refined results. See [here](http://intermine.readthedocs.org/en/latest/webapp/linking-in/) for the complete list of how you can link in to your mine. - We've added the ontology term identifier to the last column of enrichment widget downloads for better analysis - Client library now works with Python 3 - #848 - On report pages, you can now [collapse collections](http://intermine.readthedocs.org/en/latest/webapp/report-page/report-page/#references-collections) by default and [expand templates](http://intermine.readthedocs.org/en/latest/webapp/report-page/report-page/#templates) by default. - #777 - List conversion - config ignored, all fields displayed - #721 - Allow adding of collections to top summary section of report - #839 - Missing icons in Results tables - #838 - Remove filter icon not being displayed - #716 - Biopax data download - script broken - #714 - PrecomputedTableManager fails with SimpleObjects - #712 - Prevent precomputed tables being generated unneccesarily - #681 - Error with Java 8 - #676 - Duplicate queries being run with LOOKUP constraints and wildcards - #664 - Fix auto-generated python code from List analysis page when logged in - #653 - Error precomputing objectstore queries containing COUNT(*) - #648 - Write parser for GVF files - #631 - Remove modmine from core code - #626 - Results tables - Integer values appear in results table but not in summary popup - #625 - choosing either ruby or python in the API tag logs an error - #617 - URL for docs is wrong in generated code - #614 - Report page - if bad ID don't throw exception - #604 - Move model out of OS package - #603 - Keyword search - memory not freed when removing webapp - #584 - Generated JSON omitting optional constraints - #580 - make sure users don't need our CDN - #575 - Export to galaxy fails - #574 - Galaxy export failing - #570 - Keyword search - select all checkbox doesn't work - #568 - Precompute fails with outer joins - #567 - Update to new file format - FlyBase expression - #538 - icons on results not working in firefox / IE 11 - #529 - Label decoration of faceItem.value in keywordSearchResults.jsp - #504 - results tables: Item check boxes not updated correctly - #470 - OAuth2 authentication - #465 - Use equals instead of LIKE when user has selected EQUALS - #452 - LOOKUP query with wildcard is slow - #449 - Quick search results only retrieving first 100 rows - #415 - Too many segments in a range query kills the webapp - #273 - Some Permalinks are only temporary - #54 - Query for empty collections / references

See here for the complete list of tickets: https://github.com/intermine/intermine/issues?q=is%3Aissue+milestone%3A%22InterMine+1.4%22+is%3Aclosed