SBML Test Suite v3.4.0

[Lucian Smith]

The SBML Team is pleased to announce the release of version 3.4.0 of the SBML Test Suite:

https://github.com/sbmlteam/sbml-test-suite/releases/tag/3.4.0

This release collects a variety of updates to the tests from the past several years, and updates all three branches of the test suite.

For the 'semantic' tests:
* Tests 1781-1821 added, including tests for:
  * Default values of various elements.
  * Local parameters shadowing reaction IDs and species reference IDs.
  * Case sensitivity.
  * 0D-compartments in Level 3.
  * As suggested in https://github.com/sbmlteam/sbml-test-suite/issues/72: tests where model elements are given mathematical literal IDs.
* A number of previously-existing tests updated, including:
  * Correct 'comp' model files with extra id's.
  * Correct 'comp' models with external model definitions.
  * Adjust some mathematical tests so they test the function in question and not the encoding of integers as doubles.
  * Add L2v5 and L3v2 models to all relevant tests that didn't already have them.

For the 'syntactic' tests:
* A number of new tests added, particularly for the spatial package.  (Now up to date as of libsbml 5.19.3)  **These tests are attached as a binary for this release, and are not present in the source release.**

For the 'stochastic' tests:
* New tests 40-100, which test implementation of the 'distrib' package functions.

The SBML Test Suite Runner has not been updated in this release, just the tests themselves. The SBML Test Suite Database (http://raterule.caltech.edu/Facilities/Database) of tests and test results from various tools has been updated, so new results from simulators may be uploaded.

Please contact us either via email or via the github issue tracker if you have any comments, problems, or suggestions.  Thank you!

-Lucian Smith,

on behalf of the SBML Team.

[Ilya Kiselev]

Dear Lucian, SBML team members.

I have a question regarding test 01787.

In the description it says that:

"An algebraic rule sets the value for non-constant elements in its formula.  Here, one of the two symbols used is a constant species, and the other is a species with no explicitly-set value, meaning that it defaults to constant=false.  This lets S1 change to 3, and leaves S2 fixed at 3, instead of the other way 'round.An algebraic rule is the only mathematical construct that cares (outside of validation) whether an element is constant or not"

However in the model both species S1 and S2 are not constant, rather S2 has boundaryCondition = "true".

Is it an error or did I miss something?

Looking forward to hearing from you!

Best regards,

Ilya

четверг, 20 января 2022 г. в 21:50:51 UTC+7, luciano...@gmail.com:

[Lucian Smith]

You are correct!  I'll update the test, and thanks for the report.

-Lucian

--
For questions or feedback about the sbml-discuss list,
contact sbml...@googlegroups.com
---
You received this message because you are subscribed to the Google Groups "sbml-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to sbml-discuss...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/sbml-discuss/774a1418-b2b1-4be9-97b1-833b09ec3a9cn%40googlegroups.com.

[Ilya Kiselev]

Dear Lucian!

I've checked tests again (this time for older versions of SBML) and noticed that in tests 1167 and 1168 for SBML l3v1 files "enzyme-identical-l3v1.xml" have external model definitions referring  "enzyme_model.xml" instead of "enzyme_model-3v1.xml".

Best regards, 

Ilya

воскресенье, 27 февраля 2022 г. в 03:18:21 UTC+7, luciano...@gmail.com:

[Lucian Smith]

Thank you!  This is now fixed with https://github.com/sbmlteam/sbml-test-suite/pull/89 and was merged into the 'develop' branch of the suite, so will be fixed in the next release (or you can just grab the current develop branch).

-Lucian

To view this discussion on the web visit https://groups.google.com/d/msgid/sbml-discuss/ab31e032-02c6-4b79-b875-8100e6470483n%40googlegroups.com.