Skip to main content
SpringerLink
Log in

Comparing the chemical spaces of metabolites and available chemicals: models of metabolite-likeness

  • Full Length Paper
  • Published:
Molecular Diversity Aims and scope Submit manuscript

Abstract

The chemical space covered by compounds involved in metabolic reactions was compared with that of a random dataset of purchasable compounds by chemoinformatics techniques. The comparison was based on 3D structure, 2D structure, or descriptors of global properties, by means of self-organizing maps, random forests, and classification trees. The overlap between metabolites and non-metabolites was observed to be the least in the space defined by the global descriptors, the most discriminatory features being the number of OH groups, presence of aromatic systems, and molecular weight. Discrimination between the two datasets was achieved with accuracy up to 97%. Models were built to produce a metabolite-likeness parameter. A relationship between metabolite-likeness and ready biodegradability was observed.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Log in via an institution

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Similar content being viewed by others

Abbreviations

CPG NN:

Counterpropagation neural network

CV:

Cross-validation

MITI:

Japanese Ministry of International Trade and Industry

OOB:

Out-of-bag

RDF:

Radial distribution function

RF:

Random forest

SOM:

Self-organizing map

References

  1. Kulkarni SA, Zhu J, Blechinger S (2005) In silico techniques for the study and prediction of xenobiotic metabolism: a review. Xenobiotica 35:955–973, and references therein

    Google Scholar 

  2. Ellis LBM, Roe D, Wackett LP (2006) The University of Minnesota biocatalysis/biodegradation database: the first decade. Nucleic Acids Res 34:D517–D521

    Article  CAS  Google Scholar 

  3. Holliday GL, Bartlett GJ, Almonacid DE, O’Boyle NM, Murray-Rust P, Thornton JM, Mitchell JBO (2005) MACiE: a database of enzyme reaction mechanisms. Bioinformatics 21:4315–4316

    Article  CAS  Google Scholar 

  4. Reitz M, Sacher O, Tarkhov A, Trümbach D, Gasteiger J (2004) Enabling the exploration of biochemical pathways. Org Biomol Chem 2:3226–3237

    Article  CAS  Google Scholar 

  5. Hattori M, Okuno Y, Goto S, Kanehisa M (2003) Development of a chemical structure comparison method for integrated analysis of chemical and genomic information in the metabolic pathways. J Am Chem Soc 125:11853–11865

    Article  CAS  Google Scholar 

  6. Nobeli I, Ponstingl H, Krissinel EB, Thornton JM (2003) A structure-based anatomy of E. coli metabolome. J Mol Biol 334:697–719

    CAS  Google Scholar 

  7. Latino DARS, Aires-de-Sousa J (2006) Genome-scale classification of metabolic reactions: a chemoinformatics approach. Angew Chem Int Ed 45:2066–2069

    Article  CAS  Google Scholar 

  8. Kanehisa M, Goto S, Kawashima S, Okuno Y, Hattori M (2004) The KEGG resource for deciphering the genome. Nucleic Acids Res 32:D277–D280

    Article  CAS  Google Scholar 

  9. Kanehisa M, Goto S, Hattori M, Aoki-Kinoshita KF, Itoh M, Kawashima S, Katayama T, Araki M, Hirakawa M (2006) From genomics to chemical genomics: new developments in KEGG. Nucleic Acids Res 34:D354–D357

    Article  CAS  Google Scholar 

  10. Irwin JJ, Shoichet BK (2005) ZINC—a free database of commercially available compounds for virtual screening. J Chem Inf Comput Sci 45:177–182

    CAS  Google Scholar 

  11. Henkel T, Brunne RM, Muller H, Reichel F (1999) Statistical investigation into the structural complementarity of natural products and synthetic compounds. Angew Chem Int Ed Engl 38:643–647

    Article  CAS  Google Scholar 

  12. Stahura FL, Jeffrey WG, Xue L, Bajorath J (2000) Distinguishing between natural products and synthetic molecules by descriptor Shannon entropy analysis and binary QSAR calculations. J Chem Inf Comput Sci 40:1245–1252

    Article  CAS  Google Scholar 

  13. Lee M-L, Schneider G (2001) Scaffold architecture and pharmacophoric properties of natural products and trade drugs: application in the design of natural product-based combinatorial libraries. J Comb Chem 3:284–289

    Article  CAS  Google Scholar 

  14. Feher M, Schmidt JM (2003) Property distributions: differences between drugs, natural products, and molecules from combinatorial chemistry. J Chem Inf Comput Sci 43:218–227

    Article  CAS  Google Scholar 

  15. Koch MA, Schuffenhauer A, Scheck M, Wetzel S, Casaulta M, Odermatt A, Ertl P, Waldmann H (2005) Charting biologically relevant chemical space: a structural classification of natural products (SCONP). Proc Natl Acad Sci USA 102:17272–17277

    Article  CAS  Google Scholar 

  16. Karakoc E, Sahinalp SC, Cherkasov A (2006) Comparative QSAR and fragments distribution analysis of drugs, druglikes, metabolic substances, and antimicrobial compounds. J Chem Inf Model 46:2167–2182

    Article  CAS  Google Scholar 

  17. Cherkasov A (2006) Can ‹bacterial-metabolite-likeness’ model improve odds of ‹in silico’ antibiotic discovery? J Chem Inf Model 46:1214–1222

    CAS  Google Scholar 

  18. Kohonen T (1989) Self-organization and associative memory, 3rd edn. Springer, Berlin Heidelberg New York

    Google Scholar 

  19. Breiman L (2001) Random forests. Mach Learn 45:5–32

    Article  Google Scholar 

  20. Breiman L, Friedman JH, Olshen RA, Stone CJ (2000) Classification and regression trees. Chapman & Hall/CRC, Boca Raton

    Google Scholar 

  21. Hemmer MC, Steinhauer V, Gasteiger J (1999) The prediction of the 3D structure of organic molecules from their infrared spectra. Vib Spectrosc 19:151–164

    Article  CAS  Google Scholar 

  22. Organisation for Economic Co-operation and Development. http://www.oecd.org

  23. PETRA can be tested on the website http://www2.chemie.uni-erlangen.de and is developed by Molecular Networks GmbH (Erlangen, Germany). http://www.mol-net.de

  24. National Institute of Technology and Evaluation, Japan. http://www.safe.nite.go.jp/english/db.html

  25. CORINA software is available from Molecular Networks GmbH (Erlangen, Germany). http://www.mol-net.de

  26. Aires-de-Sousa J (2002) Java tools for neural networks. Chemometrics Intell Lab Syst 61:167–173

    Article  CAS  Google Scholar 

  27. The JATOON applets are available at http://www.dq.fct.unl.pt/staff/jas/jatoon

  28. Svetnik V, Liaw A, Tong C, Culberson JC, Sheridan RP, Feuston BP (2003) Random forest: a classification and regression tool for compound classification and QSAR modeling. J Chem Inf Comput Sci 43:1947–1958

    Article  CAS  Google Scholar 

  29. R Development Core Team (2004) R: a language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. ISBN 3-900051-07-0. http://www.R-project.org

  30. Fortran original by Breiman L, Cutler A, R port by Liaw A, Wiener M (2004)

  31. Zupan J, Gasteiger J (1999) Neural networks in chemistry and drug design, 2nd edn. Wiley-VCH, Weinheim

  32. Dobson CM (2004) Chemical space and biology. Nature 432:824–828

    Article  CAS  Google Scholar 

Download references

Acknowledgments

S. Gupta acknowledges Fundação para a Ciência e Tecnologia (Lisbon, Portugal) for the postdoctoral grant SFRH/BPD/14475/2003 co-funded by the POCI 2010 EU program. Molecular Networks GmbH (Erlangen, Germany) is acknowledged for access to PETRA and CORINA software packages. The authors thank Dr Robert Boethling for assistance with the MITI data.

Author information

Authors and Affiliations

  1. REQUIMTE, CQFB, Departamento de Química, Faculdade de Ciências e Tecnologia, Universidade Nova de Lisboa, 2829-516, Caparica, Portugal

    Sunil Gupta & João Aires-de-Sousa

Authors
  1. Sunil Gupta
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. João Aires-de-Sousa
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to João Aires-de-Sousa.

Appendix

Appendix

Table 3 Confusion matrices for discrimination between metabolites and non-metabolites using the models obtained from RDF and 25 global molecular descriptors
Full size table
Fig. 7
figure 7

The importance of global molecular descriptors obtained from RF for the discrimination between metabolites and non-metabolites. (a) Mean decrease in accuracy. (b) Gini criterion. A1 number of atoms, A2 number of bonds, A3 molecular weight, A4 number of aromatic atoms, A5 mean molecular polarizability, A6 number of NH groups, A7 number of NH2 groups, A8 number of non-hydrogen atoms, A9 number of OH groups, A10 number of O atoms, A11 number of N atoms, A12 number of F atoms, A13 number of P atoms, A14 number of S atoms, A15 number of Cl atoms, A16 number of Br atoms, A17 number of I atoms, A18 total number of halogen atoms, A19 minimum partial atomic charge, A20 maximum partial atomic charge, A21 minimum charge on H, A22 maximum charge on H, A23 aromatic delocalization energy, A24 ring strain energy, A25 maximum length of C chain

Full size image
Fig. 8
figure 8

The complete classification tree derived with CART algorithm to distinguish between metabolites and non-metabolites. A3 molecular weight, A4 number of aromatic atoms, A9 number of OH groups, A20 maximum partial atomic charge, Met metabolite, NMet non-metabolite

Full size image

Rights and permissions

Reprints and permissions

About this article

Cite this article

Gupta, S., Aires-de-Sousa, J. Comparing the chemical spaces of metabolites and available chemicals: models of metabolite-likeness. Mol Divers 11, 23–36 (2007). https://doi.org/10.1007/s11030-006-9054-0

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s11030-006-9054-0

Keywords

Search

Navigation