Ninja to the rescue

Don't you hate all these long compile times? I've head of Ninja a long time ago, but never gave it much thought. In case you don't feel like following the page, Ninja is a replacement for make / nmake. So today I gave it a try:

Installation

Couldn't be simpler. I cloned the git repository, opened a VS command line, called bootstrap.py with python 2.7, and added the directory to the path.

Configuring libSBML

I figured libSBML makes for a great test project, at least on my machine a build of it with tests takes forever. So I configured a new build of libsbml, release, static runtime, with tests. Thus the configuration line is:

cmake -G Ninja -DWITH_CHECK=ON -DWITH_STATIC_RUNTIME=ON -DLIBSBML_DEPENDENCY_DIR=..\..\vs11_dependencies_32_static-release ..\..

and the build is ready to go. To collect timing information, i run using time:

\cygwin\bin\time.exe ninja

And *drumroll*, time reported: 0.00user 0.00system 2:29.59elapsed 0%CPU. If you know Windows C++ builds (on non SSD drives), you know they take forever, so having all these projects build in two and a half minutes is amazing.

counter example

Just so you know, here the counter example, of using NMake as distributed by VS11. Here the configure line is:

cmake -G "NMake Makefiles"  -DWITH_CHECK=ON -DWITH_STATIC_RUNTIME=ON -DLIBSBML_DEPENDENCY_DIR=..\..\vs11_dependencies_32_static-release -DCMAKE_BUILD_TYPE=Release ..\..

running timed build again leaves gives us: 0.00user 0.00system 8:31.14elapsed 0%CPU. A whole 6 minutes more!

More data

So I couldn’t stop there, a compilation of libSBML on OSX  took (again with tests) 1m55.769 (vs. 4m4.568 for the regular make, and 2m2.698 for make –j 4). With universal binaries (i386 + x86_64) that time doubles as expected. Not too much of a difference. But this is made up by the fact that the output of ninja displays all critical things like warnings an errors nicely.

Today I used ninja for COPASI. Just as example: CopasiSE + CopasiUI + java bindings builds in just 4 minutes, 28! As compared to the 24m28 that nmake was using.

It seems obvious that using Ninja definitely saves a bunch of time, especially when compiling on Windows. The next thing to try is to use it as external build tool in VS!

Flux Balance Tools

For the Friday morning here a collection of links to software applications and databases all around flux balance analysis.

Tools

• Systems Biology Workbench: While SBW does not in and of itself deal with flux balance analysis, it does make it easy to construct networks. And makes it easy to send models between applications.
  Download 
• FluxBalance; An application to add constraints and objectives to SBML models and to evaluate them using LPsolve.
  Download
• CBMPy: comprehensive python package for analyzing flux balance models. Installation instructions are here.
• OpenCOBRA: MATLAB toolbox for genome scale reconstruction. Instructions on how to use are here.

WEB

• F.A.M.E: Web based environment for Flux Analysis. With a tutorial here. 

Databases

• KEGG Pathway: KEGG PATHWAY Database (used by F.A.M.E for visualization).
• BIGG Database: Database of genome scale reconstructed networks. Among them the global reconstruction of the human metabolic network.

SED-ML Web Tools & COMBINE archive

A couple of days I've published the COMBINE archive project to github. Today, I'm proud to let you know, that you can now upload / download COMBINE archive also to the SED-ML web tools:

  http://sysbioapps.dyndns.org/SED-ML_Web_Tools/

If you rather try it offline on your Windows box, try the SED-ML Script editor:

  SetupSedML-win32-1.11.exe

COMBINE archive … an implementation

It has been quite some time ago, that the COMBINE archive was proposed it seemed a simple enough format. I had a library for some time, but was not yet sure where to read it. As with any format, if no files are around there will be no applications written to support them.

In any case, since the conversion is going again, I’ve decided to put all the information about the library, and an first version of a graphical frontend for it. As with any experimental frontend it is written again for Windows. The links below have the details:

• The library and all information about it,
• the source code
• the windows installer

About the User Interface

When you open an archive in the application, you will find the following layout:

On the left you find the list of files contained in the archive, separated by their type, currently only a limited number of the COMBINE standards is recognized and an even smaller number can be interacted with, but that will change in the future. When you select a file, the systems preview handler will be used to generate a preview, as you can see in the screenshot with the PDF. Also recognized are images, and websites. Can the format not be determined, the file will be displayed as plain text, which may or may not be helpful.

If an SBML file is selected, it will be displayed as plain text. But if you have SBW installed, you will see this pane on the top:

That will allow you to open the file immediately for simulation / editing with one of the defined favorite applications. Or you could choose any other installed SBW Analyzer. The next version will allow you to simulate SED-ML files, and view SBGN files … stay tunes! I’m also working on an online version for sysbioapps.

libSBML 5.8.0 Released

Just as follow up to the release of the SBML Level 3 Package for Flux Balance Constraints: we have now released a new version of libSBML, that includes the FBC package enabled in all binaries. That is not just the core library, but also all language bindings (including MATLAB). Particularly I want to highlight three features:

• the first one is full validation: all validation rules have been implemented in libSBML and will be validated automatically on a checkConsistency() call on SBMLDocuments.
• the other one are the converters: we have included converters from the COBRA annotation to the FBC package and vice versa. Examples are available online, or in the source archives including the package.
• Finally, C++ documentation of the package is available online. While package documentation is not yet available for language bindings, they do use the same function calls, so the information is transferable.

See also the full release announcement for more information.

SBML Level 3 Package for Flux Balance Constraints

It is official, the SBML Level 3 Package for Flux Balance Constraints has been released! In its first version it allows to efficiently store constraints of Flux Balance models along with objectives to be met.

You can find the full specification online under:

http://identifiers.org/combine.specifications/sbml.level-3.version-1.fbc.version-1.release-1

As of today the next release of libSBML (5.8.0) has gone into code freeze, that will include the FBC package, along with full validation and conversions from SBML Levels that use the COBRA annotations to SBML Level 3 models with the FBC package. This new version is bound to be released by the end of the month. In the meantime nightly windows builds can be found on our build server.

With this release I’ve also made a new version of my FluxBalance tool available. Version 1.7 now already uses the COBRA importer, and has been redesigned to efficiently compute flux balance solutions for even genome scale models (albeit the loading of these models still takes a while).

A big thanks to the SBML editors and the community to make this possible!

New SBML Test Suite Database!

As you might have noticed there is a new online service available on sbml.org, the SBML Test Suite Database. We of course had the SBML Test Suite for quite some time, with an amazing number of 1196 tests (to date), that now also include tests for the Flux Balance Package and even Hierarchical Modeling. This new application lets you browse all these tests, filter for tests (for example: Show me all tests, that test ‘RandomEventExecution’) and  download a selection of tests right there.

At the heart of course is the capability to *upload* a zipped set of test results, and analyze them for yourself in a private area, and once you are happy with them have them published publicly for the world to see. I’ve am working on a new version of SBW, that uses the 5.7.0 version of libSBML. And here is what a result for RoadRunner would look like:

As you can see in yellow, RoadRunner will still not support AlgebraicRules, FastReactions or CSymbolDelay. However, libSBML will flatten comp models for RoadRunner, and so all comp tests that do not use the above mentioned elements will work just fine.