A Tree Based Method for the Rapid Screening of Chemical Fingerprints

Thomas G. Kristensen, Jesper Nielsen and Christian N. S. Pedersen

Illustration of Multibit tree This is the home page supplementing the article A Tree Based Method for the Rapid Screening of Chemical Fingerprints. In the article we describe a novel method for indexing large amounts of chemical fingerprints (bit strings) for rapid searching using the Tanimoto coefficient as a similarity measure. The following items are available for download.

Source code

The source code of our implementation, including ant build script, JUnit tests and an implementation of the methods as a web service. CDK is required to compile and run the main program. The code has been tested with version 1.0.4. To run the automated tests, JUnit is required. We have used version 4.5. The web service requires a Tomcat server.

Download the source code

Compiled version

The compiled version contains all libraries necessary to run the main methods, and to use our data structures.

Download the compiled version

Full test bed

All the experiments from the article can be replicated simply by downloading the full test bed, and running the file runExperiments.sh. The file includes the necessary libraries and all the test files for running all experiments.

Download the full test bed

Valid XHTML 1.1