Enabling systems biology: Automated elucidation of metabolite structures

CINF 57

Christoph Steinbeck, c.steinbeck@uni-koeln.de, Miguel Rojas, Tobias Helmus, Egon Willighagen, and Stefan Kuhn. Research Group for Molecular Informatics, Cologne University Bioinformatics Center (CUBIC), Zuelpicher Str. 47, D-50674 Cologne, Germany
Identification and structure elucidation of unknown metabolite structures based on their spectroscopic properties form the basis for successful metabolome simulations. In a process known as dereplication, a scientist would record molecular fingerprint spectra and search spectral databases to check whether the compound at hand is already known. Only if this search in unsuccessful, it is reasonable to reach for one of the more sophisticated ab-inito tools for computer-assisted structure elucidation. Here we describe the use of free software, the easy access provided by the World Wide Web and the collaborative potential of the Open-Source movement to build a completely transparent system for computer-assisted structure elucidation and identification. Methods for the prediction of mass and NMR spectra have been developed and used as part of a fitness function in our structure elucidation systems based on stochastic chemical space generators.