The master branch of MX now features support for two very important cheminformatics capabilities: Extensible Fingerprints. Lets you fine-tune the performance of your binary path-based fingerprints. Change path depth, number of bits, and the characters used to generate the fingerprints. The default implementation differentiates cycle paths from non-cycle paths, as well as unsaturated from saturated atom types.
The most recent C&EN contains this tidbit - and this time it’s no joke: It’s not totally clear what “rotated and condensed” means, but if it’s what I’m thinking, ACS might as well stop the printing presses now. I’ve run across one or two of these “rotated and condensed” journals - and they are barely usable. The “rotated” part means that holding the thing feels like you’re flipping through a calendar.
Chris Steinbeck has an interesting post on the CDK code review process that discusses a new VF implementation. In it, he notes: For the unfamiliar, VF is a subgraph matching algorithm that has been shown to perform better than Ullmann for small graphs, and much better than Ullmann for large graphs. Faced with essentially the same problem of implementing VF in Java for MX, I abandoned my early efforts to port the VFlib C++ implementation.
Update June 18, 2009: The MX public repository has been moved to http://github.com/metamolecular/mx. Issue tracking is a critical feature for many software projects. Over the weekend I had a chance to play with a new GitHub feature - integrated issue tracking. A specific example from MX, the lightweight cheminformatics toolkit, illustrates the process. It was found that MX was throwing an exception where none should be thrown.
Ein Fehler ist aufgetreten. Sieh dir dieses Video auf www.youtube.com an oder aktiviere JavaScript, falls es in deinem Browser deaktiviert sein sollte. More like business casual, but still worth watching.
The first beta version of MX 1.0 has been released. MX is a lightweight cheminformatics toolkit written in Java that emphasizes efficiency, modularity, and readability. Incorporating many changes, this release introduces support for query atoms and Superatoms.
The molfile specification is chemistry’s de facto standard for the exchange of chemical structures. Tracing its history back to at least 1979, the V2000 Molfile format is now supported in varying degrees by nearly every vendor and product on the market today. Despite its limitations, molfile has stood the test of time remarkably well.
If you’ve ever needed to deploy Java applets, you may have found the information on best practices scattered, inconsistent, and unreliable. Applet-Fu is a small JavaScript library that makes it easy to deploy applets by taking advantage of Web standards and features already supported by mainstream browsers. More detailed discussion of Applet-Fu is available from the Metamolecular Products Blog and the ChemWriter Best Practices Deployment Page.
REST is an architectural style for Web applications that emphasizes regularity and discoverability. Unlike most other approaches, REST exploits the full power of HTTP, both in terms of expression and performance optimization. This article explores REST from the perspective of cheminformatics Web services, using the specific example of substructure search.
If you’ve ever created Web documents containing source code, you may have wondered how to easily add syntax highlighting. Over at the Metamolecular Products Blog, we needed source code highlighting and decided to find the best solution available. Depth-First uses a highlighter that’s installed as a server-side plugin.
