Advertisement

Applied Intelligence

, Volume 12, Issue 1–2, pp 117–147 | Cite as

Application of Cascade Correlation Networks for Structures to Chemistry

  • Anna Maria Bianucci
  • Alessio Micheli
  • Alessandro Sperduti
  • Antonina Starita
Article

Abstract

We present the application of Cascade Correlation for structures to QSPR (quantitative structure-property relationships) and QSAR (quantitative structure-activity relationships) analysis. Cascade Correlation for structures is a neural network model recently proposed for the processing of structured data. This allows the direct treatment of chemical compounds as labeled trees, which constitutes a novel approach to QSPR/QSAR. We report the results obtained for QSPR on Alkanes (predicting the boiling point) and QSAR of a class of Benzodiazepines. Our approach compares favorably versus the traditional QSAR treatment based on equations and it is competitive with ‘ad hoc’ MLPs for the QSPR problem.

Cascade Correlation networks constructive algorithms gradient descent QSPR QSAR 

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
    D.S. Touretzky, “Boltzcons: Dynamic symbol structures in a connectionist network,” Artificial Intellicence, vol. 46, pp. 5–46, 1990.Google Scholar
  2. 2.
    R. Rosenfeld and D.S. Touretzky, “Four capacity models for coarse-coded symbol memories,” Technical Report CMU-CS–87–182, Carnegie Mellon, 1987.Google Scholar
  3. 3.
    J.B. Pollack, “Recursive distributed representations,” Artificial Intelligence, vol. 46, no. 1/2, pp. 77–106, 1990.Google Scholar
  4. 4.
    G.E. Hinton, “Mapping part-whole hierarchies into connectionist network,” Artificial Intelligence, vol. 46, pp. 47–75, 1990.Google Scholar
  5. 5.
    T.A. Plate, “Holographic reduced representations,” IEEE Transactions on Neural Networks, vol. 6, no. 3, pp. 623–641, 1995.Google Scholar
  6. 6.
    P. Smolensky, “Tensor product variable binding and the representation of symbolic structures in connectionist systems,” Artificial Intelligence, vol. 46, pp. 159–216, 1990.Google Scholar
  7. 7.
    A. Sperduti and A. Starita, “Supervised neural networks for the classification of structures,” IEEE Transactions on Neural Networks, vol. 8, no. 3, pp. 714–735, 1997.Google Scholar
  8. 8.
    P. Frasconi, M. Gori, and A. Sperduti, “Aframework for adaptive data structures processing,” IEEE Transactions on Neural Networks, vol. 9, no. 5, pp. 768–786, 1998.Google Scholar
  9. 9.
    C. Hansch, P.P. Maloney, T. Fujita, and R.M. Muir, Nature, vol. 194, pp. 178–180, 1962.Google Scholar
  10. 10.
    C. Hansch and T. Fujita, J. Am. Chem. Soc., vol. 86, pp. 1616–1626, 1964.Google Scholar
  11. 11.
    Dimitra Hadjipavlou-Litina and Corwin Hansch, “Quantitative structure-activity relationships of the benzodiazepines.Areview and reevaluation, Chemical Reviews, vol. 94 no. 6, pp. 1483–1505, 1994.Google Scholar
  12. 12.
    L.H. Hall and L.B. Kier, “The molecular connectivity chi indexes and kappa shape indexes in structure-property modeling,” in Reviews in Computational Chemistry, VCH Publishers, Inc., New York, ch. 9, pp. 367–422. 1991.Google Scholar
  13. 13.
    J. Zupan and J. Gasteiger, Neural Networks for Chemists: An introduction, VCH Publishers: NY(USA), 1993.Google Scholar
  14. 14.
    Ajay, “A unified framework for using neural networks to build QSARs,” J. Med. Chem., vol. 36, pp. 3565–3571, 1993.Google Scholar
  15. 15.
    D. Cherqaoui and D. Villemin, “Use of neural network to determine the boiling point of alkanes,” J. Chem. Soc. Faraday Trans., vol. 90, no. 1, pp. 97–102, 1994.Google Scholar
  16. 16.
    C. Goller, “A Connectionist Approach for Learning Search-Control Heuristics for Automated Deduction Systems,” Ph.D. Thesis, Technical University Munich, Computer Science, 1997.Google Scholar
  17. 17.
    A. Sperduti, D. Majidi, and A. Starita, “Extended cascadecorrelation for syntactic and structural pattern recognition,” in Advances in Structural and Syntactical Pattern Recognition, edited by P. Perner, P.Wang, and A. Rosenfeld, Springer-Verlag, Berlin, pp. 90–99, 1996. Lecture notes in Computer Science, vol. 1121.Google Scholar
  18. 18.
    H. Kubinyi, Burger's Medicinal Chemistry and Drug Discovery, vol. 1, fifth edn., John Wiley & Sons, Inc: New York, pp. 528–530, 1995.Google Scholar
  19. 19.
    H. Kubinyi and U. Abraham, “Practical problems in PLS analyses,” in 3D-QSAR in Drug Design. Theory Methods and Application, edited by H. Kubinyi, ESCOM: Leiden, 1993, pp. 717–728.Google Scholar
  20. 20.
    D. Rogers and A.J. Hopfinger, “Application of genetic function approximation to quantitative structure-activity relationships and quantitative-property relationships,” J. Chem. Inf. Comput. Sci., vol. 34, no. 4, pp. 854–866, 1993.Google Scholar
  21. 21.
    S.S. So and M. Karplus, “Evolutionary optimization in quantitative structure-activity relationship: An application of genetic neural networks,” J. Med. Chem., vol. 39, pp. 1521–1530, 1996.Google Scholar
  22. 22.
    D.H. Rouvray, “Should we have designs on topological indices ?,” in Chemical Applications of Topology and Graph Theory, edited by R.B. King, Elsevier Science Publishing Company, pp. 159–177, 1983.Google Scholar
  23. 23.
    J.A. Burns and G.M. Whitesides, “Feed-forward neural networks in chemistry: Mathematical system for classification and pattern recognition,” Chemical Reviews, vol. 93, no. 8, pp. 2583–2601, 1993.Google Scholar
  24. 24.
    D.H. Rouvray, Computational Chemical Graph Theory, Nova Science Publishers: New York, p. 9, 1990.Google Scholar
  25. 25.
    M. Barysz, G. Jashari, R.S. Lall, V.K. Srivastava, and N. Trinajstic, “On the distance matrix of molecules containing heteroatoms,” in Chemical Applications of Topology and Graph Theory, edited by R.B. King, Elsevier Science Publishing Company, pp. 222–230, 1983.Google Scholar
  26. 26.
    V.R. Magnuson, D.K. Harris, and S.C. Basak, “Topological indices based on neighborhood symmetry: Chemical and biological application,” in Chemical Applications of Topology and Graph Theory, edited by R.B. King, Elsevier Science Publishing Company, pp. 178–191, 1983.Google Scholar
  27. 27.
    Y. Suzuki, T. Aoyama, and H. Ichikawa, “Neural networks applied to quantitative structure-activity relationships,” J. Med. Chem., vol. 33, pp. 2583–2590, 1990.Google Scholar
  28. 28.
    A.F. Duprat, T. Huynh, and G. Dreyfus, “Towards a principled methodology for neural network design and performance evaluation in QSAR; Application to the prediction of LogP,” J. Chem. Inf. Comput. Sci., vol. 38, no. 4, pp. 586–594, 1999.Google Scholar
  29. 29.
    L.K. Peterson, “Quantitative structure-activity relationships in carboquinones and benzodiazepines using counter-propagation neural networks,” J. Chem. Inf. Comput. Sci., vol. 35, no. 5, pp. 896–904, 1995.Google Scholar
  30. 30.
    Shuhui Liu, Ruisheng Zhang, Mancang Liu, and Zhide Hu, “Neural networks-topological indices approach to the prediction of properties of alkene,” J. Chem. Inf. Comput. Sci., vol. 37, pp. 1146–1151, 1997.Google Scholar
  31. 31.
    D.W. Elrod, G.M. Maggiora, and R.G. Trenary, “Application of neural networks in chemistry, 1. Prediction of electrophilic aromatic substitution reactions,” J. Chem. Inf. Comput. Sci., vol. 30, pp. 447–484, 1990.Google Scholar
  32. 32.
    V. Kvasnička and J. Pospichal, “Application of neural networks in chemistry.prediction of product distribution of nitration in a series of monosubstituted benzenes,” J. Mol. Struct. (Theochem), vol. 235, pp. 227–242, 1991.Google Scholar
  33. 33.
    S.E. Fahlman and C. Lebiere, “The cascade-correlation learning architecture,” in Advances in Neural Information Processing Systems 2, edited by D.S. Touretzky, San Mateo, CA: Morgan Kaufmann, pp. 524–532, 1990.Google Scholar
  34. 34.
    S.E. Fahlman, “The recurrent cascade-correlation architecture,” Technical Report CMU-CS–91–100, Carnegie Mellon, 1991.Google Scholar
  35. 35.
    S. Muggleton and L. De Raedt, “Inductive logic programming: Theory and methods,” Journal of Logic Programming, vol. 19, 20, pp. 629–679, 1994.Google Scholar
  36. 36.
    A.M. Bianucci, A. Micheli, A. Sperduti, and A. Starita, “Quantitative structure-activity relationships of benzodiazepines by Recursive Cascade Correlation,” in Proceedings of IJCNN '98-IEEE World Congress on Computational Intelligence, Anchorage, Alaska, May 1998, pp. 117–122.Google Scholar

Copyright information

© Kluwer Academic Publishers 2000

Authors and Affiliations

  • Anna Maria Bianucci
    • 1
  • Alessio Micheli
    • 2
  • Alessandro Sperduti
    • 2
  • Antonina Starita
    • 2
  1. 1.Dipartimento di Scienze FarmaceutichePisaItaly
  2. 2.Dipartimento di InformaticaUniversità di PisaPisaItaly

Personalised recommendations