Pharmaceutical Research

, Volume 26, Issue 9, pp 2216–2224 | Cite as

Predicting Inhibitors of Acetylcholinesterase by Regression and Classification Machine Learning Approaches with Combinations of Molecular Descriptors

Research Paper



Acetylcholinesterase (AChE) is both a therapeutic target for Alzheimer’s disease and a target for organophosphorus, carbamates and chemical warfare agents. Prediction of the likelihood of compounds interacting with this enzyme is therefore important from both therapeutic and toxicological perspectives.

Materials and Methods

Support vector machine classification and regression models with molecular descriptors derived from Shape Signatures and the Molecular Operating Environment (MOE) application software were built and tested using a set of piperidine AChE inhibitors (N = 110).


The combination of the alignment free Shape Signatures and 2D MOE descriptors with the Support Vector Regression method outperforms the models based solely on 2D and internal 3D (i3D) MOE descriptors, and is comparable with the best previously reported PLS model based on CoMFA molecular descriptors (\( {\text{r}}_{\text{test,SVR}}^2 = 0.48 \) vs. \( {\text{r}}_{\text{test,PLS}}^2 = 0.47 \) from Sutherland et al. J Med Chem 47:5541–5554, 2004). Support Vector Classification algorithms proved superior to a classifier based on scores from the molecular docking program GOLD, with the overall prediction accuracies being QSVC(10CV) = 74% and QSVC(LNO) = 67% vs. QGOLD = 56%.


These new machine learning models with combined descriptor schemes may find utility for predicting novel AChE inhibitors.


acetylcholinesterase docking machine learning molecular operating environment quantitative structure activity relationship shape signatures descriptors support vector classification support vector machine support vector regression 



We gratefully acknowledge all those involved in the application and development of Shape Signatures. Support for this work has been provided by the USEPA-funded Environmental Bioinformatics and Computational Toxicology Center (ebCTC), under STAR Grant number GAD R 832721-010, and by the Defense Threat Reduction Agency, under contract number HDTRA- BB07TAS020. This work was also funded by NIH -GM081394 from the National Institute of General Medical Sciences (to WJW). This work has not been reviewed by and does not represent the opinions of the funding agencies.

Supplementary material

11095_2009_9937_MOESM1_ESM.doc (36 kb)
Supplemental Table 1 Data from Sutherland et al. (9). (DOC 36 kb)
11095_2009_9937_MOESM2_ESM.doc (37 kb)
Supplemental Table 2 SVM Classification of 110 AChE compounds from Sutherland et al. (9). The dividing boundary was set at IC50 = 100 nM resulting in 51 strong and 59 weak inhibitors. (DOC 37 kb)
11095_2009_9937_MOESM3_ESM.doc (36 kb)
Supplemental Table 3 SVM classification of 110 AChE compounds from Sutherland et al. (9). The dividing boundary was set at IC50 = 250 nM resulting in 65 strong and 45 weak inhibitors. (DOC 37 kb)
11095_2009_9937_MOESM4_ESM.doc (36 kb)
Supplemental Table 4 SVM classification of 110 AChE compounds from Sutherland et al. (9). Dividing boundary was set at IC50 = 500 nM resulting in 73 strong and 37 weak inhibitors. (DOC 37 kb)


  1. 1.
    Moretto A. Experimental and clinical toxicology of anticholinesterase agents. Toxicol Lett. 1998;102–103:509–13.PubMedCrossRefGoogle Scholar
  2. 2.
    Castro A, Martinez A. Peripheral and dual binding site acetylcholinesterase inhibitors: implications in treatment of Alzheimer’s disease. Mini Rev Med Chem. 2001;1:267–72.PubMedCrossRefGoogle Scholar
  3. 3.
    Barril X, Orozco M, Luque FJ. Towards improved acetylcholinesterase inhibitors: a structural and computational approach. Mini Rev Med Chem. 2001;1:255–66.PubMedCrossRefGoogle Scholar
  4. 4.
    Kaur J, Zhang MQ. Molecular modelling and QSAR of reversible acetylcholines-terase inhibitors. Curr Med Chem. 2000;7:273–94.PubMedGoogle Scholar
  5. 5.
    Cramer RD, Patterson DE, Bunce JD. Comparative Molecular Field Analysis (CoMFA). 1. Effect of shape on binding of steroids to carrier proteins. J Am Chem Soc. 1988;110:5959–67.CrossRefGoogle Scholar
  6. 6.
    Tong W, Collantes ER, Chen Y, Welsh WJ. A comparative molecular field analysis study of N-benzylpiperidines as acetylcholinesterase inhibitors. J Med Chem. 1996;39:380–7.PubMedCrossRefGoogle Scholar
  7. 7.
    Golbraikh A, Bernard P, Chretien JR. Validation of protein-based alignment in 3D quantitative structure-activity relationships with CoMFA models. Eur J Med Chem. 2000;35:123–36.PubMedCrossRefGoogle Scholar
  8. 8.
    El Yazal J, Rao SN, Mehl A, Slikker W Jr. Prediction of organophosphorus acetylcholinesterase inhibition using three-dimensional quantitative structure-activity relationship (3D-QSAR) methods. Toxicol Sci. 2001;63:223–32.PubMedCrossRefGoogle Scholar
  9. 9.
    Sutherland JJ, O’Brien LA, Weaver DF. A comparison of methods for modeling quantitative structure-activity relationships. J Med Chem. 2004;47:5541–54.PubMedCrossRefGoogle Scholar
  10. 10.
    Fernandez M, Caballero J. Ensembles of Bayesian-regularized genetic neural networks for modeling of acetylcholinesterase inhibition by huprines. Chem Biol Drug Des. 2006;68:201–12.PubMedCrossRefGoogle Scholar
  11. 11.
    Akula N, Lecanu L, Greeson J, Papadopoulos V. 3D QSAR studies of AChE inhibitors based on molecular docking scores and CoMFA. Bioorg Med Chem Lett. 2006;16:6277–80.PubMedCrossRefGoogle Scholar
  12. 12.
    Jung M, Tak J, Lee Y, Jung Y. Quantitative structure-activity relationship (QSAR) of tacrine derivatives against acetylcholinesterase (AChE) activity using variable selections. Bioorg Med Chem Lett. 2007;17:1082–90.PubMedCrossRefGoogle Scholar
  13. 13.
    Manchester J, Czermiński R. SAMFA: simplifying molecular descriptors for 3D-QSAR. J Chem Inf Model. 2008;48:1167–73.PubMedCrossRefGoogle Scholar
  14. 14.
    Chekmarev DS, Kholodovych V, Balakin KV, Ivanenkov Y, Ekins S, Welsh WJ. Shape signatures: new descriptors for predicting cardiotoxicity in silico. Chem Res Toxicol. 2008;21:1304–14.PubMedCrossRefGoogle Scholar
  15. 15.
    Kortagere S, Chekmarev D, Welsh WJ, Ekins S. New predictive models for blood-brain barrier permeability of drug-like molecules. Pharm Res. 2008;25:1836–45.PubMedCrossRefGoogle Scholar
  16. 16.
    Jones G, Willett P, Glen RC, Leach AR, Taylor R. Development and validation of a genetic algorithm for flexible docking. J Mol Biol. 1997;267:727–48.PubMedCrossRefGoogle Scholar
  17. 17.
    Gasteiger J, Marsili M. Iterative partial equalization of orbital electronegativity–a rapid access to atomic charges. Tetrahedron 1980;36:3219–28.CrossRefGoogle Scholar
  18. 18.
    Zauhar RJ, Moyna G, Tian L, Li Z, Welsh WJ. Shape signatures: a new approach to computer-aided ligand- and receptor-based drug design. J Med Chem. 2003;46:5674–90.PubMedCrossRefGoogle Scholar
  19. 19.
    Nagarajan K, Zauhar R, Welsh WJ. Enrichment of ligands for the serotonin receptor using the Shape Signatures approach. J Chem Inf Model. 2005;45:49–57.PubMedCrossRefGoogle Scholar
  20. 20.
    Kortagere S, Chekmarev D, Welsh WJ, Ekins S. Hybrid scoring and classification approaches to predict human pregnane X receptor activators. Pharm Res. 2009;26(4):1001-11.PubMedCrossRefGoogle Scholar
  21. 21.
    Wang CY, Ai N, Arora S, Erenrich E, Nagarajan K, Zauhar R, et al. Identification of previously unrecognized antiestrogenic chemicals using a novel virtual screening approach. Chem Res Toxicol. 2006;19:1595–601.PubMedCrossRefGoogle Scholar
  22. 22.
    Meek PJ, Liu Z, Tian L, Wang CY, Welsh WJ, Zauhar RJ. Shape Signatures: speeding up computer aided drug discovery. Drug Discov Today. 2006;11:895–904.PubMedCrossRefGoogle Scholar
  23. 23.
    Kortagere S, Welsh WJ. Development and application of hybrid structure based method for efficient screening of ligands binding to G-protein coupled receptors. J Comput Aided Mol Des. 2006;20:789–802.PubMedCrossRefGoogle Scholar
  24. 24.
    Whitley DC, Ford MG, Livingstone DJ. Unsupervised forward selection: a method for eliminating redundant variables. J Chem Inf Comput Sci. 2000;40:1160–8.PubMedGoogle Scholar
  25. 25.
    Geladi P, Kowalski B. Partial least-squares:a tutorial. Anal Chim Acta. 1986;185:1–17.CrossRefGoogle Scholar
  26. 26.
    Cortes C, Vapnik V. Support vector networks. Machine Learn. 1995;20:273–93.Google Scholar
  27. 27.
    Vapnik V. Statistical learning theory. New York: Wiley; 1998.Google Scholar
  28. 28.
    Kecman V. Learning and soft computing: support vector machines, neural networks and Fuzzy logic models. Cambridge: MIT; 2001.Google Scholar
  29. 29.
    Ivanciuc O. Application of support vector machines in chemistry. Rev Comp Chem. 2007;23:291–400.CrossRefGoogle Scholar
  30. 30.
    Chen YZ, editor. Current QSAR techniques for toxicology. Hoboken: Wiley; 2007.Google Scholar
  31. 31.
    Xue Y, Yap CW, Sun LZ, Cao ZW, Wang JF, Chen YZ. Prediction of P-glycoprotein substrates by a support vector machine approach. J Chem Inf Comput Sci. 2004;44:1497–505.PubMedGoogle Scholar
  32. 32.
    Leong MK. A novel approach using pharmacophore ensemble/support vector machine (PhE/SVM) for prediction of hERG liability. Chem Res Toxicol. 2007;20:217–26.PubMedCrossRefGoogle Scholar
  33. 33.
    Ung CY, Li H, Yap CW, Chen YZ. In silico prediction of pregnane X receptor activators by machine learning approaches. Mol Pharmacol. 2007;71:158–68.PubMedCrossRefGoogle Scholar
  34. 34.
    Song M, Breneman C, Bi J, Sukumar N, Bennett K, Cramer S, et al. Prediction of protein retention times in anion-exchange chromatography systems using support vector regression. J Chem Inf Compu Sci. 2002;42:1347–57.Google Scholar
  35. 35.
    Yap CW, Li ZR, Chen YZ. Quantitative structure-pharmacokinetic relationships for drug clearance by using statistical learning methods. J Mol Graph Model. 2006;24:383–95.PubMedCrossRefGoogle Scholar
  36. 36.
    Chang CC, Lin CJ. LIBSVM: a library for support vector machines, 2001.
  37. 37.
    Matthews BW. Comparison of the predicted and observed secondary structure of T4 phage lysozyme. Biochim Biophys Acta. 1975;405:442–51.PubMedGoogle Scholar
  38. 38.
    Kryger G, Harel M, Giles K, Toker L, Velan B, Lazar A, et al. Structures of recombinant native and E202Q mutant human acetylcholinesterase complexed with the snake-venom toxin fasciculin-II. Acta Crystallogr Sect D. 2000;56:1385–94.CrossRefGoogle Scholar
  39. 39.
    Guo J, Hurley MH, Wright JB, Lushington GH. A docking score function for estimating ligand-protein interactions: application to acetylcholinesterase inhibition. J Med Chem. 2004;47:5492–500.PubMedCrossRefGoogle Scholar
  40. 40.
    Ekins S, Embrechts MJ, Breneman CM, Jim K, Wery J-P. Novel applications of Kernel-partial least squares to modeling a comprehensive array of properties for drug discovery. In: Ekins S, editor. Computational toxicology: risk assessment for pharmaceutical and environmental chemicals. Hoboken: Wiley-Interscience; 2007. p. 403–32.Google Scholar
  41. 41.
    Todeschini R, Consonni V. Handbook of molecular descriptors. Weinheim: Wiley-VCH; 2000.CrossRefGoogle Scholar

Copyright information

© Springer Science+Business Media, LLC 2009

Authors and Affiliations

  1. 1.Department of Pharmacology and Environmental Bioinformatics & Computational Toxicology Center (ebCTC)University of Medicine & Dentistry of New Jersey, Robert Wood Johnson Medical SchoolPiscatawayUSA
  2. 2.Collaborations in ChemistryJenkintownUSA
  3. 3.Department of Pharmaceutical SciencesUniversity of MarylandCollege ParkUSA
  4. 4.Department of Microbiology and ImmunologyDrexel University College of MedicinePhiladelphiaUSA

Personalised recommendations