Review

ChemSpaceShuttle: A tool for data mining in drug discovery by classification, projection, and 3D visualization

  1. Alireza Givehchi1,5,
  2. Axel Dietrich1,2,
  3. Paul Wrede1,3,
  4. Gisbert Schneider4

Article first published online: 10 JUL 2003

DOI: 10.1002/qsar.200310007

Issue

QSAR & Combinatorial Science

QSAR & Combinatorial Science

Volume 22, Issue 5, pages 549–559, July 2003

Additional Information

How to Cite

Givehchi, A., Dietrich, A., Wrede, P. and Schneider, G. (2003), ChemSpaceShuttle: A tool for data mining in drug discovery by classification, projection, and 3D visualization. QSAR & Combinatorial Science, 22: 549–559. doi: 10.1002/qsar.200310007

Author Information

  1. 1

    CallistoGen AG, Neuendorfstr. 24b, D-16761 Hennigsdorf, Germany

  2. 2

    Present address: The Genetics Company, Wagistraße 27, CH-8952 Schlieren, Switzerland

  3. 3

    Present address: Freie Universität Berlin, Institut für Molekularbiologie und Bioinformatik, Arnimallee 22, D-14195 Berlin, Germany

  4. 4

    Johann Wolfgang Goethe-Universität, Institut für Organische Chemie und Chemische Biologie, Marie-Curie-Str. 11, D-60439 Frankfurt, Germany

  5. 5

    Johann Wolfgang Goethe-Universität, Institut für Organische Chemie und Chemische Biologie, Marie-Curie-Str. 11, D-60439 Frankfurt, Germany, Tel: +49 – (0)69 798 29824, Fax: +49 – (0)69 798 29826

Publication History

  1. Issue published online: 8 DEC 2003
  2. Article first published online: 10 JUL 2003
  3. Manuscript Accepted: 23 DEC 2002
  4. Manuscript Received: 12 NOV 2002

SEARCH

SEARCH BY CITATION

ARTICLE TOOLS

Get PDF (414K)

Keywords:

  • Drug discovery / Library design / Neural network / Self-organizing map / Principal component

Abstract

The ChemSpaceShuttle toolbox provides a graphical interface allowing for ligand-based design of focused compound libraries by means of linear and non-linear projection techniques and clustering algorithms. The software implements a non-linear encoder network and non-linear partial least squares for the projection of high-dimensional descriptor vectors into a three dimensional space for visualisation. Compound clustering by a self-organising map (SOM) is incorporated. Visualization can facilitate the selection of compounds with desired properties from large compound libraries. Two sample applications are presented: similarity-based compound selection for focused library design, and classification of drugs and nondrugs. A version of ChemSpaceShuttle is freely available at URL: http://gecco.org.chemie.uni-frankfurt.de/gecco.html.

Get PDF (414K)