Advertisement

Neural Networks in Building QSAR Models

  • Igor I. Baskin
  • Vladimir A. Palyulin
  • Nikolai S. Zefirov
Part of the Methods in Molecular Biology™ book series (MIMB, volume 458)

Abstract

This chapter critically reviews some of the important methods being used for building quantitative structure-activity relationship (QSAR) models using the artificial neural networks (ANNs). It attends predominantly to the use of multilayer ANNs in the regression analysis of structure-activity data. The highlighted topics cover the approximating ability of ANNs, the interpretability of the resulting models, the issues of generalization and memorization, the problems of overfitting and overtraining, the learning dynamics, regularization, and the use of neural network ensembles. The next part of the chapter focuses attention on the use of descriptors. It reviews different descriptor selection and preprocessing techniques; considers the use of the substituent, substructural, and superstructural descriptors in building common QSAR models; the use of molecular field descriptors in three-dimensional QSAR studies; along with the prospects of “direct” graph-based QSAR analysis. The chapter starts with a short historical survey of the main milestones in this area.

Keywords

Artificial neural networks QSAR back propagation learning generalization 

References

  1. 1.
    Hiller SA, Glaz AB, Rastrigin LA, Rosenblit AB (1971) Recognition of physiological activity of chemical compounds on perceptron with random adaptation of structure. Dokl Akad Nauk SSSR 199:851–853.Google Scholar
  2. 2.
    Minsky M, Papert S (1969) Perseptrons. MIT Press, Cambridge, MA.Google Scholar
  3. 3.
    Hiller SA, Golender VE, Rosenblit AB, Rastrigin, LA, Glaz AB (1973) Cybernetic methods of drug design. I. Statement of the problem—the perceptron approach. Comp. Biomed. Res 6:411–421.Google Scholar
  4. 4.
    Aoyama T, Suzuki Y, Ichikawa H (1990) Neural networks applied to structure-activity relationships. J Med Chem 33:905–908.PubMedGoogle Scholar
  5. 5.
    Aoyama T, Suzuki,Y, Ichikawa H (1990) Neural networks applied to pharmaceutical problems. III. Neural networks applied to quantitative structure-activity relationship (QSAR) analysis. J Med Chem 33:2583–2590.PubMedGoogle Scholar
  6. 6.
    Zupan J, Gasteiger J (1999) Neural networks in chemistry. Wiley-VCH, Weinheim.Google Scholar
  7. 7.
    Devillers J (ed) (1996) Neural networks in QSAR and drug design. Academic Press, San Diego, CA.Google Scholar
  8. 8.
    Manallack DT, Livingstone DJ (1995) Neural networks and expert systems in molecular design. Methods and Principles in Medicinal Chemistry 3:293–318.Google Scholar
  9. 9.
    Winkler DA, Maddalena DJ (1995) QSAR and neural networks in life sciences. Series in Mathematical Biology and Medicine 5:126–163.Google Scholar
  10. 10.
    Maddalena DJ (1996) Applications of artificial neural networks to quantitative structure-activity relationships. Expert Opinion on Therapeutic Patents 6:239–251.Google Scholar
  11. 11.
    Anzali S, Gasteiger J, Holzgrabe U, Polanski J, Sadowski J, Teckentrup A, Wagener M (1998) The use of self-organizing neural networks in drug design. Persp Drug Disc Des 9-11:273–299.Google Scholar
  12. 12.
    Schneider G, Wrede P (1998) Artificial neural networks for computer-based molecular design. Progress in Biophysics and Molecular Biology 70:175–222.PubMedGoogle Scholar
  13. 13.
    Kovesdi I, Dominguez-Rodriguez MF, Orfi L, Naray-Szabo G, Varro A, Papp JG, Matyus P (1999) Application of neural networks in structure-activity relationships. Med Res Rev 19:249–269.PubMedGoogle Scholar
  14. 14.
    Manallack DT, Livingstone DJ (1999) Neural networks in drug discovery: have they lived up to their promise? Eur J Med Chem 34:195–208.Google Scholar
  15. 15.
    Ochoa C, Chana A, Stud M (2001) Applications of neural networks in the medicinal chemistry field. Curr Med Chem 1:247–256.Google Scholar
  16. 16.
    Terfloth L, Gasteiger J (2001) Neural networks and genetic algorithms in drug design. Drug Discovery Today 6:S102–S108.Google Scholar
  17. 17.
    Winkler DA, Burden FR (2002) Application of neural networks to large dataset QSAR, virtual screening, and library design. Methods in Mol Biol 201:325–367.Google Scholar
  18. 18.
    Halberstam NM, Baskin I I, Palyulin VA, Zefirov NS (2003) Neural networks as a method for elucidating structure-property relationships for organic compounds. Russ Chem Rev 72:629–649.Google Scholar
  19. 19.
    Kaiser KLE (2003) The use of neural networks in QSARs for acute aquatic toxicological endpoints. J Mol Struct (Theochem) 622:85–95.Google Scholar
  20. 20.
    Kaiser KLE (2003) Neural networks for effect prediction in environmental and health issues using large datasets. QSAR Comb Sci 22:185–190.Google Scholar
  21. 21.
    Livingstone D J, Manallack DT (2003) Neural networks in 3D QSAR. QSAR Comb Sci 22:510–518.Google Scholar
  22. 22.
    Niculescu SP (2003) Artificial neural networks and genetic algorithms in QSAR. J Mol Struct (Theochem) 622:71–83.Google Scholar
  23. 23.
    Novic M, Vracko M (2003) Artificial neural networks in molecular structures-property studies. Data Handling in Science and Technology 23:231–256.Google Scholar
  24. 24.
    Polanski J (2003) Self-organizing neural networks for pharmacophore mapping. Advanced Drug Delivery Reviews 55:1149–1162.PubMedGoogle Scholar
  25. 25.
    Taskinen J, Yliruusi J (2003) Prediction of physicochemical properties based on neural network modeling. Advanced Drug Delivery Reviews 55:1163–1183.PubMedGoogle Scholar
  26. 26.
    Winkler DA (2004) Neural networks as robust tools in drug lead discovery and development. Molecular Biotechnology 27:138–167.Google Scholar
  27. 27.
    Winkler DA, Burden FR (2004) Bayesian neural nets for modeling in drug discovery. Drug Discovery Today 2:104–111.Google Scholar
  28. 28.
    Shoji R (2005) The potential performance of artificial neural networks in QSTRs for predicting ecotoxicity of environmental pollutants. Curr. Comput-Aided Drug Des 1:65–72.Google Scholar
  29. 29.
    Livingstone DJ, Hesketh G, Clayworth D (1991) Novel method for the display of multivariate data using neural networks. J Mol Graph 9:115–118.PubMedGoogle Scholar
  30. 30.
    Rose VS, Croall IF, MacFie HJH (1991) An application of unsupervised neural network methodology (Kohonen topology-preserving mapping) to QSAR analysis. QSAR 10:6–15.Google Scholar
  31. 31.
    Peterson KL (1992) Counter-propagation neural networks in the modeling and prediction of Kovats indexes for substituted phenols. Anal Chem 64:379–386.Google Scholar
  32. 32.
    Liu Q, Hirono S, Moriguchi I (1992) Comparison of the functional-link net and the generalized delta rule net in quantitative structure-activity relationship studies. Chem Pharm Bull 40:2962–2969.Google Scholar
  33. 33.
    Baskin II, Palyulin VA, Zefirov NS (1993) Methodology of searching for direct correlations between structures and properties of organic compounds by using computational neural networks. Dokl Akad Nauk 333:176–179.Google Scholar
  34. 34.
    Baskin II, Palyulin VA, Zefirov NS (1997) A neural device for searching direct correlations between structures and properties of chemical compounds. J Chem Inf Comput Sci 37:715–721.Google Scholar
  35. 35.
    Lohninger H (1993) Evaluation of neural networks based on radial basis functions and their application to the prediction of boiling points from structural parameters. J Chem Inf Comput Sci 33:736–744.Google Scholar
  36. 36.
    Brinn M, Walsh PT, Payne MP, Bott B (1993) Neural network classification of mutagens using structural fragment data. SAR QSAR Env Res 1:169–210.Google Scholar
  37. 37.
    Tetko IV, Luik AI, Poda GI. (1993) Applications of neural networks in structure-activity relationships of a small number of molecules. J Med Chem 36:811–814.PubMedGoogle Scholar
  38. 38.
    Sutter JM, Dixon SL, Jurs PC (1995) Automated descriptor selection for quantitative structure-activity relationships using generalized simulated annealing. J Chem Inf Comput Sci 35:77–84.Google Scholar
  39. 39.
    So S-S, Karplus M (1996) Evolutionary optimization in quantitative structure-activity relationship: an application of genetic neural networks. J Med Chem 39:1521–1530.PubMedGoogle Scholar
  40. 40.
    Viswanadhan VN, Mueller GA, Basak SC, Weinstein JN (1996) A new QSAR algorithm combining principal component analysis with a neural network: application to calcium channel antagonists. Network Science 2.Google Scholar
  41. 41.
    Domine D, Devillers J, Wienke D, Buydens L (1997) ART 2-A for optimal test series design in QSAR. J Chem Inf Comput Sci 37:10–17.Google Scholar
  42. 42.
    Sato K, Nakagawa J, Matsuzaki H (1997) Bayesian neural network approach to quantitative structure-activity relationships in carboquinones. Ann Rept Tohoku College of Pharmacy 44:187–193.Google Scholar
  43. 43.
    Kaiser KLE, Niculescu SP, McKinnon MB (1997) On simple linear regression, multiple linear regression, and elementary probabilistic neural network with Gaussian kernel's performance in modeling toxicity values to fathead minnow based on Microtox data, octanol/water partition coefficient, and various structural descriptors for a 419-compound dataset. In: Quantitative structure-activity relationships in environmental sciences, VII, Proceedings of QSAR 96, Elsinore, Denmark, June 24–28, 1996, pp. 285–297.Google Scholar
  44. 44.
    Polanski J (1997) The receptor-like neural network for modeling corticosteroid and testosterone binding globulins. J Chem Inf Comput Sci 37:553–561.PubMedGoogle Scholar
  45. 45.
    Kovalishyn VV, Tetko IV, Luik AI, Kholodovych VV, Villa AEP, Livingstone DJ (1998) Neural network studies 3. Variable selection in the cascade-correlation learning architecture. J Chem Inf Comput Sci 38:651–659.Google Scholar
  46. 46.
    Zupan J, Novic M (1999) Optimization of structure representation for QSAR studies. Anal Chim Acta 388:243–250.Google Scholar
  47. 47.
    Espinosa G, Yaffe D, Cohen Y, Arenas A, Giralt F (2000) Neural network based quantitative structural property relations (QSPRs) for predicting boiling points of aliphatic hydrocarbons. J Chem Inf Comput Sci 40:859–879.PubMedGoogle Scholar
  48. 48.
    Li, P, Cheng Y-Y (2000) Studies on quantitative structure-activity relationships of benzodiazepines using fuzzy neural networks. Gaodeng Xuexiao Huaxue Xuebao 21:1473–1478.Google Scholar
  49. 49.
    Micheli A, Sperduti A, Starita A, Bianucci AM. (2001) Analysis of the internal representations developed by neural networks for structures applied to quantitative structure-activity relationship studies of benzodiazepines. J Chem Inf Comput Sci 41:202–218.PubMedGoogle Scholar
  50. 50.
    Tetko IV, Kovalishyn VV, Livingstone DJ (2001) Volume learning algorithm artificial neural networks for 3D QSAR studies. J Med Chem 44:2411–2420.PubMedGoogle Scholar
  51. 51.
    Polanski J, Gieleciak R, Bak A (2002) The comparative molecular surface analysis (CoMSA)—a nongrid 3D QSAR method by a coupled neural network and PLS system: predicting pK(a) values of benzoic and alkanoic acids. J Chem Inf Comput Sci 42:184–191.PubMedGoogle Scholar
  52. 52.
    Patankar SJ, Jurs PC (2002) Prediction of glycine/NMDA receptor antagonist inhibition from molecular structure. J Chem Inf Comput Sci 42:1053–1068.PubMedGoogle Scholar
  53. 53.
    Mosier PD, Jurs PC (2002) QSAR/QSPR studies using probabilistic neural networks and generalized regression neural networks. J Chem Inf Comput Sci 42:1460–1470.PubMedGoogle Scholar
  54. 54.
    Espinosa G, Arenas A, Giralt F (2002) An integrated SOM-fuzzy ARTMAP neural system for the evaluation of toxicity. J Chem Inf Comput Sci 42:343–359.PubMedGoogle Scholar
  55. 55.
    Agrafiotis DK, Cedeno W (2002) Feature selection for structure-activity correlation using binary particle swarms. J Med Chem 45:1098–1107.PubMedGoogle Scholar
  56. 56.
    Izrailev S, Agrafiotis DK (2002) Variable selection for QSAR by artificial ant colony systems. SAR QSAR Env Res 13:417–423.Google Scholar
  57. 57.
    Arakawa M, Hasegawa K, Funatsu K (2003) Application of the novel molecular alignment method using the Hopfield neural network to 3D-QSAR. J Chem Inf Comput Sci 43:1396–1402.PubMedGoogle Scholar
  58. 58.
    Hemmateenejad B, Akhond M, Miri R, Shamsipur M (2003) Genetic algorithm applied to the selection of factors in principal component-artificial neural networks: application to QSAR study of calcium channel antagonist activity of 1,4-dihydropyridines (nifedipine analogs). J Chem Inf Comput Sci 43:1328–1334.PubMedGoogle Scholar
  59. 59.
    Baskin II, Halberstam NM, Artemenko NV, Palyulin VA, Zefirov NS (2003) NASAWIN—a universal software for QSPR/QSAR studies. In: Ford M et al. (eds) EuroQSAR 2002 designing drugs and crop protectants: processes, problems and solutions. Blackwell Publishing, pp. 260–263.Google Scholar
  60. 60.
    Cedeno W, Agrafiotis DK (2003) Application of niching particle swarms to QSAR and QSPR. In: Ford M et al. (eds) EuroQSAR 2002 designing drugs and crop protectants: processes, problems and solutions. Blackwell Publishing, pp. 255–259.Google Scholar
  61. 61.
    Bayram E, Santago P II, Harris R, Xiao Y-D, Clauset AJ, Schmitt JD (2004) Genetic algorithms and self-organizing maps: a powerful combination for modeling complex QSAR and QSPR problems. J Comp-Aided Mol Des 18:483–493.Google Scholar
  62. 62.
    Baurin N, Mozziconacci J-C, Arnoult E, Chavatte P, Marot C, Morin-Allory L (2004) 2D QSAR consensus prediction for high-throughput virtual screening. An application to COX-2 inhibition modeling and screening of the NCI. J Chem Inf Comput Sci 44:276–285.PubMedGoogle Scholar
  63. 63.
    Maggiora GM, Elrod DW, Trenary RG (1992) Computational neural networks as model-free mapping devices. J Chem Inf Comput Sci 32:732–741.Google Scholar
  64. 64.
    Kolmogorov AN (1957) On the representation of continuous functions of many variables by superposition of continuous functions of one variable and addition. Dokl Akad Nauk SSSR 114:953–956.Google Scholar
  65. 65.
    Kurková V (1992) Kolmogorov's theorem and multilayer neural networks. Neural Networks 5:501–506.Google Scholar
  66. 66.
    Manallack DT, Livingstone DJ (1994) Limitations of functional-link nets as applied to QSAR data analysis. Quantitative Structure-Activity Relationships 13:18–21.Google Scholar
  67. 67.
    Stuetzle W, Mittal Y (1979) Some comments on the asymptotic behavior of robust smoothers. In: Gasser T, Rosenblatt M (eds) Smoothing techniques for curve estimation. Springer-Verlag, Helderberg.Google Scholar
  68. 68.
    Tetko IV (2002) Neural network studies, 4. Introduction to associative neural networks. J Chem Inf Comput Sci 42:717–728.PubMedGoogle Scholar
  69. 69.
    Raevsky OA, Trepalin SV, Trepalina HP, Gerasimenko VA, Raevskaja OE (2002) SLIPPER-2001—software for predicting molecular properties on the basis of physicochemical descriptors and structural similarity. J Chem Inf Comput Sci 42:540–549.PubMedGoogle Scholar
  70. 70.
    Féraud R, Clérot F (2002) A methodology to explain neural network classification. Neural Networks 15:237–246.PubMedGoogle Scholar
  71. 71.
    Guha R, Stanton DT, Jurs PC (2005) Interpreting computational neural network quantitative structure-activity relationship models: a detailed interpretation of the weights and biases. J Chem Inf Mod 45:1109–1121.Google Scholar
  72. 72.
    Aoyama, T, Ichikawa H (1992) Neural networks as nonlinear structure-activity relationship analyzers. Useful functions of the partial derivative method in multilayer neural networks. J Chem Inf Comput Sci 32:492–500.Google Scholar
  73. 73.
    Baskin II, Ait AO, Halberstam NM, Palyulin VA, Zefirov NS (2002) An approach to the interpretation of backpropagation neural network models in QSAR studies. SAR QSAR Env Res 13:35–41.Google Scholar
  74. 74.
    Halberstam NM, Baskin II, Palyulin VA, Zefirov NS (2002) Quantitative structure-conditions-property relationship studies. Neural network modeling of the acid hydrolysis of esters. Mendeleev Commun 12:185–186.Google Scholar
  75. 75.
    Guha R, Jurs PC (2005) Interpreting computational neural network QSAR models: a measure of descriptor importance. J Chem Inf Mod 45:800–806.Google Scholar
  76. 76.
    Stanton DT (2003) On the physical interpretation of QSAR models. J Chem Inf Comput Sci 43:1423–1433.PubMedGoogle Scholar
  77. 77.
    Manallack D, Livingstone DJ (1992) Artificial neural networks: application and chance effects for QSAR data analysis. Med Chem Res 2:181–190.Google Scholar
  78. 78.
    Livingstone DJ, Salt DW (1992) Regression analysis for QSAR using neural networks. Bioorg Med Chem Let 2:213–218.Google Scholar
  79. 79.
    Livingstone DJ, Manallack DT (1993) Statistics using neural networks: chance effects. J Med Chem 36:1295–1297.PubMedGoogle Scholar
  80. 80.
    Tetko IV, Livingstone DJ, Luik AI (1995) Neural network studies, 1. Comparison of overfitting and overtraining. J Chem Inf Comput Sci 35:826–833.Google Scholar
  81. 81.
    Andrea TA, Kalayeh H (1991) Applications of neural networks in quantitative structure-activity relationships of dihydrofolate reductase inhibitors. J Med Chem 34:2824–2836.PubMedGoogle Scholar
  82. 82.
    Ezhov AA, Shumsky SA (1998) Neurocomputing and its applications in economics and business. MIPhI, Moscow.Google Scholar
  83. 83.
    Rissanen J (1983) A universal prior for the integers and estimation by minimum description length. Annals of Statistics 11:416–431.Google Scholar
  84. 84.
    Tikhonov AN, Arsenin VY (1977) Solutions of ill-posed problems. Winston & Sons, Washington, DC.Google Scholar
  85. 85.
    Burden FR, Winkler DA (1999) Robust QSAR models using Bayesian regularized neural networks. J Med Chem 42:3183–3187.PubMedGoogle Scholar
  86. 86.
    Agrafiotis DK, Cedeño W, Lobanov VS (2002) On the use of neural network ensembles in QSAR and QSPR. J Chem Inf Comput Sci 42:903–911.PubMedGoogle Scholar
  87. 87.
    Whitley DC, Ford MG, Livingstone DJ (2000) Unsupervised forward selection: a method for eliminating redundant variables. J Chem Inf Comput Sci 40:1160–1168.PubMedGoogle Scholar
  88. 88.
    Tetko IV, Villa AEP, Livingstone DJ (1996) Neural network studies, 2. Variable Selection. J Chem Inf Comput Sci 36:794–803.PubMedGoogle Scholar
  89. 89.
    Artemenko NV, Baskin II, Palyulin VA, Zefirov NS (2003) Artificial neural network and fragmental approach in prediction of physicochemical properties of organic compounds. Russ Chem Bull 52:20–29.Google Scholar
  90. 90.
    Gustafsson MG. (2005) Independent component analysis yields chemically interpretable latent variables in multivariate regression. J Chem Inf Comput Sci 45:1244–1255.Google Scholar
  91. 91.
    Hansch C, Fujita T (1964) σ-ρ-π Analysis. A method for the correlation of biological activity and chemical structure. J Am Chem Soc 86:1616–1626.Google Scholar
  92. 92.
    Free SM Jr, Wilson JW (1964) A mathematical contribution to structure-activity studies. J Med Chem 7:395–399.PubMedGoogle Scholar
  93. 93.
    Kvasnička V, Sklenák Š, Pospichal J (1993) Neural network classification of inductive and resonance effects of substituents. J Am Chem Soc 115:1495–1500.Google Scholar
  94. 94.
    Baskin II, Keschtova SV, Palyulin VA, Zefirov NS (2000) Combining molecular modeling with the use of artificial neural networks as an approach to predicting substituent constants and bioactivity. In: Gundertofte K, J?rgensen FS (eds) Molecular modeling and prediction of bioactivity. Kluwer Academic/Plenum Publishers, New York, pp. 468–469.Google Scholar
  95. 95.
    Chiu T-L, So S-S (2004) Development of neural network QSPR models for Hansch substituent constants, 1. Method and validations. J Chem Inf Comput Sci 44:147–153.PubMedGoogle Scholar
  96. 96.
    Chiu T-L, So S-S (2004) Development of neural network QSPR models for Hansch substituent constants, 2. Applications in QSAR studies of HIV-1 reverse transcriptase and dihydrofolate reductase inhibitors. J Chem Inf Comput Sci 44:154–160.PubMedGoogle Scholar
  97. 97.
    Baskin II, Halberstam NM, Mukhina TV, Palyulin VA, Zefirov NS (2001) The learned symmetry concept in revealing quantitative structure-activity relationships with artificial neural networks. SAR QSAR Env Res 12:401–416.Google Scholar
  98. 98.
    Baskin II, Skvortsova MI, Stankevich IV, Zefirov NS (1994) Basis of invariants of labeled molecular graphs. Doklady Chemistry 339:231–234.Google Scholar
  99. 99.
    Baskin II, Skvortsova MI, Stankevich IV, Zefirov NS (1995) On the basis of invariants of labeled molecular graphs. J Chem Inf Comput Sci 35:527–531.Google Scholar
  100. 100.
    Baskin II, Palyulin VA, Zefirov NS (1993) Computational neural networks as an alternative to linear regression analysis in studies of quantitative structure-property relationships for the case of the physicochemical properties of hydrocarbons. Dokl Akad Nauk 332:713–716.Google Scholar
  101. 101.
    Artemenko NV, Baskin II, Palyulin VA, Zefirov NS (2001) Prediction of physical properties of organic compounds by artificial neural networks in the framework of substructural approach. Doklady Chemistry 381:317–320.Google Scholar
  102. 102.
    Artemenko NV, Palyulin VA, Zefirov NS (2002) Neural-network model of the lipophilicity of organic compounds based on fragment descriptors. Doklady Chemistry 383:114–116.Google Scholar
  103. 103.
    Zhokhova NI, Baskin II, Palyulin VA, Zefirov AN, Zefirov NS (2003) Fragmental descriptors in QSPR: flash point calculations. Russ Chem Bull 52:1885–1892.Google Scholar
  104. 104.
    Zhokhova NI, Baskin II, Palyulin VA, Zefirov AN, Zefirov NS (2003) Calculation of the enthalpy of sublimation by the QSPR method with the use of a fragment approach. Russ J Appl Chem 76:1914–1919.Google Scholar
  105. 105.
    Zhokhova NI, Baskin II, Palyulin VA, Zefirov AN, Zefirov NS (2004) Fragment descriptors in QSPR: application to magnetic susceptibility calculations. J Structural Chem 45:626–635.Google Scholar
  106. 106.
    Liubimova IK, Abilev SK, Gal'berstam NM, Baskin II, Paliulin VA, Zefirov NS (2001) Computer-aided prediction of the mutagenic activity substituted polycyclic compounds. Biology Bull 28:139–145.Google Scholar
  107. 107.
    Radchenko EV, Baranova OD, Palyulin VA, Zefirov NS (2003) Non-linear molecular field topology analysis by means of artificial neural networks. In: Ford M et al. (eds) EuroQSAR 2002 designing drugs and crop protectants: processes, problems and solutions. Blackwell Publishing, pp. 317–318.Google Scholar
  108. 108.
    Cramer RD, Patterson DE, Bunce JD (1988) Comparative molecular field analysis (CoMFA), 1. Effect of shape and binding of steroids to carrier proteins. J Am Chem Soc 110:5959–5967.Google Scholar
  109. 109.
    Hasegawa K, Matsuoka S, Arakawa M, Funatsu K (2002) New molecular surface-based 3D-QSAR method using Kohonen neural network and three-way PLS. Comput Chem 26:583–589.PubMedGoogle Scholar
  110. 110.
    Baskin II, Tikhonova IG, Palyulin VA, Zefirov NS (2006) A combined approach to building 3D-QSAR models. In: Proc 15th European symposium on quantitative structure-activity relationships & molecular modeling, September 5–10 2004, Istanbul, Turkey (in press).Google Scholar
  111. 111.
    Kireev DB (1995) ChemNet: a novel neural network based method for graph/property mapping. J Chem Inf Comput Sci 35:175–180.Google Scholar
  112. 112.
    Ivanciuc O (2001) Molecular structure encoding into artificial neural networks topology. Roumanian Chem Quart Rev 8:197–220.Google Scholar
  113. 113.
    Baldi P, Rosen-Zvi M (2005) On the relationship between deterministic and probabilistic directed graphical models: from Bayesian networks to recursive neural networks. Neural Networks 18:1080–1086.PubMedGoogle Scholar

Copyright information

© Springer-Verlag 2006

Authors and Affiliations

  • Igor I. Baskin
    • 1
  • Vladimir A. Palyulin
    • 2
  • Nikolai S. Zefirov
    • 3
  1. 1.Department of ChemistryMoscow State UniversityMoscowRussia
  2. 2.Department of ChemistryMoscow State UniversityMoscowRussia
  3. 3.Department of ChemistryMoscow State UniversityMoscowRussia

Personalised recommendations