SteinBlog » Structure Elucidation

Open Access Journal of Cheminformatics now live!

Christoph Steinbeck — Tue, 17 Mar 2009 17:00:51 +0000

I’m delighted to announce that the first open access journal of our field, the Journal of Cheminformatics, is now live and has published its first articles. Journal of Cheminformatics is a new open access journal from Chemistry Central publishing peer-reviewed research in all aspects of cheminformatics and molecular modelling. It is run by Editors-in-Chief David J. Wild (Indiana University) and myself (European Bioinformatics Institute).
Amongst the launch articles are an Editorial by David J. Wild on Grand challenges for cheminformatics, a Commentary by Steven M. Bachrach on Chemistry publication – making the revolution and last but not least an article by Tony Williams and coauthors on Computer Assisted Structure Elucidation (CASE), one of my own fields of research.
You can view articles and submit your manuscripts at www.jcheminf.com. Please share this information with your colleagues working in the field of chemical information who may be interested in this new journal.

Computer-Assisted Structure Elucidation by 1D Carbon NMR

Christoph Steinbeck — Wed, 31 Jan 2007 10:42:00 +0000

We (my group) combined a couple of methods from various of our projects (Bioclipse, CDK, NMRShiftDB, SENECA and LUCY) to create a very simple setup of Computer-Assisted Structure Elucidation. We demonstrated it with the carbon spectrum of a-pinene InChI=1/C10H16/c1-7-4-5-8-6-9(7)10(8,2)3/h4,8-9H,5-6H2,1-3H3

The result is a Bioclipse plugin (bc_seneca) which takes a peak-picking of a measured (experimental) carbon NMR spectrum and a molecular formula as an input. It then uses the CDK deterministic structure generator, generates all possible isomers and predicts carbon shifts for all of them. The result is a list of isomers ranked by similarity of their predicted spectra with the measured one.

[[image:case.png::center:0]]

The correct solution for our test case is a-pinene shown in the upper left corner. Here is also the input CMLSpect file, which is part of the Bioclipse bc_data plugin.


type="NMR" xmlns:cmlDict="http://www.xml-cml.org/dict/cmlDict"
xmlns:subst="http://www.xml-cml.org/dict/substDict"
xmlns:siUnits="http://www.xml-cml.org/units/siUnits"
xmlns:nmr="http://www.nmrshiftdb.org/dict"
xmlns:units="http://www.xml-cml.org/units/units"
xmlns:cml="http://www.xml-cml.org/dict/cml"
xmlns:macie="http://www.xml-cml.org/dict/macie"
xmlns="http://www.xml-cml.org/schema">
  
    Unreported
    Unreported

Human Metabolome Database published – with NMR search

Christoph Steinbeck — Sun, 28 Jan 2007 15:09:00 +0000

Scientists at the University of Alberta have published a database with small molecules from human metabolism.

CBCNews has an article.

The article mentions about 2500 substances to be listed in the database. An interesting feature of this Human Metabolome Database is an NMR and MS spectrum search. Unfortunately, the search does not feature an overly flexible similarity function but after some tweaking of parameter, one can get a result. A list of carbon NMR shifts such as “10, 20, 30, 110, 120, 200″ yields results if you increase the shift tolerance to 5ppm and lower the similarity threshold to 50. Note that all of the NMR spectra you get in this result list are predicted and not measured.

Our group here at Cologne University Bioinformatics Center (CUBIC) maintains an NMRShiftDB, which could be used as a source for measured NMR spectra in this project.