Solving the Structure-Property Problem Using k-NN Classification

  • Aleksandr Perevoznikov
  • Alexey Shestov
  • Evgenii Permiakov
  • Mikhail Kumskov
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 6744)


The solution of the ”structure-property” based on the molecular graphs descriptors selection with k-NN classifier is proposed. The results of comparing the construction of predictive models using the search and without it are given. The stability of the classifier function construction quality is tested using the test sample.


Pattern Recognition QSAR QSPR k-NN 


  1. 1.
    Kohov, V.A.: The Method of Quantifying the Similarity of Graphs Based on Structural Spectrs. Izvestiya RAN, Technical Cybernetics 5, 143–159 (1994)Google Scholar
  2. 2.
    Svitanko, I.V., Devetyarov, D.A., Tcheboukov, D.E., Dolmat, M.S., Zakharov, A.M., Grigoryeva, S.S., Chichua, V.T., Ponomareva, L.A., Kumskov, M.I.: QSAR Modeling on the Basis of 3D Descriptors Representing the Electrostatic Molecular Surface (Ambergris Fragrances). Mendeleev Communications 17(2), 90–91 (2007)CrossRefGoogle Scholar
  3. 3.
    Svitanko, I.V., Kumskov, M.I., Tcheboukov, D.E., Dolmat, M.S., Zakharov, A.M., Ponomareva, L.A., Grigoryeva, S.S., Chichua, V.T.: QSAR Modeling on the Base of Electrostatic Molecular Surface (Amber Fragrances). In: 16th European Symposium on Quantum Structure-Activity Relationships and Molecular Modelling, EuroQSAR, Italy (2007)Google Scholar
  4. 4.
    Kumskov, M.I., Smolenskiy, E.A., Ponomareva, L.A., Mitushev, D.F., Zefirov, N.S.: System of structural descriptors for solving structure-property problem. Reports of the Academy of Sciences 336(1), 64–66 (1994)Google Scholar
  5. 5.
    Hurst, T., Heritage, T.: HQSAR - A Highly Predictive QSAR Technique Based on Molecular Holograms. In: 213th ACS National Meeting, San Francisco, CA, CINF 019 (1997)Google Scholar
  6. 6.
    Hoffman, B., Cho, S.J., Zheng, W., Wyrick, S., Nicols, D.E., Mailman, R.B., Tropsha, A.: Quantitative structure-activity relationship modelling of dopamine D[1] antagonists using comparative molecular field analysis, generic algorithms-partial least-squares, and K nearest neighbour methods. J. Med. Chem. 42, 3217–3226 (1999)CrossRefGoogle Scholar
  7. 7.
    Zheng, W., Tropsha, A.: Novel variable selection quantitative structure-property relationship approach based on the k-nearest-neighbour principle. J. Chem. Inf. Comput. Sci. 40, 185–194 (2000)CrossRefGoogle Scholar
  8. 8.
    Clark, M., Cramer III, R.D., Jones, D.M., Patterson, D.E., Simeroth, P.E.: Comparative Molecular Field Analysis(CoMFA) Toward Its Use with 3DStructural Databases. Tetrahedron Comput. Methodol. 3, 47–59 (1990)CrossRefGoogle Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2011

Authors and Affiliations

  • Aleksandr Perevoznikov
    • 1
  • Alexey Shestov
    • 1
  • Evgenii Permiakov
    • 1
  • Mikhail Kumskov
    • 1
  1. 1.Faculty of Mechanics and Mathematics, Department of Computational MathematicsLomonosov Moscow State UniversityMoscowRussian Federation

Personalised recommendations