The AAPS Journal

, 20:54 | Cite as

Improved Prediction of Blood–Brain Barrier Permeability Through Machine Learning with Combined Use of Molecular Property-Based Descriptors and Fingerprints

  • Yaxia Yuan
  • Fang Zheng
  • Chang-Guo Zhan
Research Article Theme: Better Drugs for Better Life: Drug Discovery and Development Colloquium 2017
Part of the following topical collections:
  1. Theme: Better Drugs for Better Life: Drug Discovery and Development Colloquium 2017


Blood–brain barrier (BBB) permeability of a compound determines whether the compound can effectively enter the brain. It is an essential property which must be accounted for in drug discovery with a target in the brain. Several computational methods have been used to predict the BBB permeability. In particular, support vector machine (SVM), which is a kernel-based machine learning method, has been used popularly in this field. For SVM training and prediction, the compounds are characterized by molecular descriptors. Some SVM models were based on the use of molecular property-based descriptors (including 1D, 2D, and 3D descriptors) or fragment-based descriptors (known as the fingerprints of a molecule). The selection of descriptors is critical for the performance of a SVM model. In this study, we aimed to develop a generally applicable new SVM model by combining all of the features of the molecular property-based descriptors and fingerprints to improve the accuracy for the BBB permeability prediction. The results indicate that our SVM model has improved accuracy compared to the currently available models of the BBB permeability prediction.


blood–brain barrier permeability molecular descriptor fingerprint physical property modeling 



The authors acknowledge the Computer Center at the University of Kentucky for supercomputing time on a Dell Supercomputer Cluster consisting of 388 nodes or 4816 processors.

Funding Information

This work was supported in part by the National Science Foundation (NSF grant CHE-1111761) and the National Institutes of Health (NIH grants UH2/UH3 DA041115, R01 DA035552, R01 DA032910, R01 DA013930, R01 DA025100, and UL1TR001998).

Compliance with Ethical Standards

Conflict of Interest

The authors declare that they have no conflict of interest.

Supplementary material

12248_2018_215_MOESM1_ESM.pdf (556 kb)
ESM 1 (PDF 555 kb)


  1. 1.
    George A. The design and molecular modeling of CNS drugs. Curr Opin Drug Discov Dev. 1999;2(4):286–92.Google Scholar
  2. 2.
    Puzyn T, Leszczynski J, Cronin MT. Recent advances in QSAR studies: methods and applications. Berlin: Springer Science & Business Media; 2010.CrossRefGoogle Scholar
  3. 3.
    Bicker J, Alves G, Fortuna A, Falcão A. Blood–brain barrier models and their relevance for a successful development of CNS drug delivery systems: a review. Eur J Pharm Biopharm. 2014;87(3):409–32.CrossRefPubMedGoogle Scholar
  4. 4.
    Abraham MH. The factors that influence permeation across the blood–brain barrier. Eur J Med Chem. 2004;39(3):235–40.CrossRefPubMedGoogle Scholar
  5. 5.
    Rose K, Hall LH, Kier LB. Modeling blood-brain barrier partitioning using the electrotopological state. J Chem Inf Comput Sci. 2002;42(3):651–66.CrossRefPubMedGoogle Scholar
  6. 6.
    Crivori P, Cruciani G, Carrupt P-A, Testa B. Predicting blood−brain barrier permeation from three-dimensional molecular structure. J Med Chem. 2000;43(11):2204–16.CrossRefPubMedGoogle Scholar
  7. 7.
    Ooms F, Weber P, Carrupt P-A, Testa B. A simple model to predict blood–brain barrier permeation from 3D molecular fields. Biochim Biophys Acta (BBA) – Mol Basis Dis. 2002;1587(2):118–25.CrossRefGoogle Scholar
  8. 8.
    Gerebtzoff G, Seelig A. In silico prediction of blood−brain barrier permeation using the calculated molecular cross-sectional area as main parameter. J Chem Inf Model. 2006;46(6):2638–50.CrossRefPubMedGoogle Scholar
  9. 9.
    Zhao YH, Abraham MH, Ibrahim A, Fish PV, Cole S, Lewis ML, et al. Predicting penetration across the blood-brain barrier from simple descriptors and fragmentation schemes. J Chem Inf Model. 2007;47(1):170–5.CrossRefPubMedGoogle Scholar
  10. 10.
    Lanevskij K, Japertas P, Didziapetris R. Improving the prediction of drug disposition in the brain. Expert Opin Drug Metab Toxicol. 2013;9(4):473–86.CrossRefPubMedGoogle Scholar
  11. 11.
    Mehdipour AR, Hamidi M. Brain drug targeting: a computational approach for overcoming blood–brain barrier. Drug Discov Today. 2009;14(21):1030–6.CrossRefPubMedGoogle Scholar
  12. 12.
    Nagpal K, Singh SK, Mishra DN. Drug targeting to brain: a systematic approach to study the factors, parameters and approaches for prediction of permeability of drugs across BBB. Expert Opin Drug Deliv. 2013;10(7):927–55.CrossRefPubMedGoogle Scholar
  13. 13.
    Di L, Kerns EH, Carter GT. Strategies to assess blood–brain barrier penetration. Expert Opin Drug Discovery. 2008;3(6):677–87.CrossRefGoogle Scholar
  14. 14.
    Vilar S, Chakrabarti M, Costanzi S. Prediction of passive blood–brain partitioning: straightforward and effective classification models based on in silico derived physicochemical descriptors. J Mol Graph Model. 2010;28(8):899–903.CrossRefPubMedPubMedCentralGoogle Scholar
  15. 15.
    Zah J, Terre'Blanche G, Erasmus E, Malan SF. Physicochemical prediction of a brain–blood distribution profile in polycyclic amines. Bioorg Med Chem. 2003;11(17):3569–78.CrossRefPubMedGoogle Scholar
  16. 16.
    Young RC, Mitchell RC, Brown TH, Ganellin CR, Griffiths R, Jones M, et al. Development of a new physicochemical model for brain penetration and its application to the design of centrally acting H2 receptor histamine antagonists. J Med Chem. 1988;31(3):656–71.CrossRefPubMedGoogle Scholar
  17. 17.
    Dagenais C, Avdeef A, Tsinman O, Dudley A, Beliveau R. P-glycoprotein deficient mouse in situ blood–brain barrier permeability and its prediction using an in combo PAMPA model. Eur J Pharm Sci. 2009;38(2):121–37.CrossRefPubMedPubMedCentralGoogle Scholar
  18. 18.
    Lanevskij K, Japertas P, Didziapetris R, Petrauskas A. Ionization-specific prediction of blood–brain permeability. J Pharm Sci. 2009;98(1):122–34.CrossRefPubMedGoogle Scholar
  19. 19.
    Lanevskij K, Dapkunas J, Juska L, Japertas P, Didziapetris R. QSAR analysis of blood–brain distribution: the influence of plasma and brain tissue binding. J Pharm Sci. 2011;100(6):2147–60.CrossRefPubMedGoogle Scholar
  20. 20.
    Luco JM. Prediction of the brain− blood distribution of a large set of drugs from structurally derived descriptors using partial least-squares (PLS) modeling. J Chem Inf Comput Sci. 1999;39(2):396–404.CrossRefPubMedGoogle Scholar
  21. 21.
    Adenot M, Lahana R. Blood-brain barrier permeation models: discriminating between potential CNS and non-CNS drugs including P-glycoprotein substrates. J Chem Inf Comput Sci. 2004;44(1):239–48.CrossRefPubMedGoogle Scholar
  22. 22.
    Mente S, Lombardo F. A recursive-partitioning model for blood–brain barrier permeation. J Comput Aided Mol Des. 2005;19(7):465–81.CrossRefPubMedGoogle Scholar
  23. 23.
    Shen J, Du Y, Zhao Y, Liu G, Tang Y. In silico prediction of blood–brain partitioning using a chemometric method called genetic algorithm based variable selection. Mol Informatics. 2008;27(6):704–17.Google Scholar
  24. 24.
    Li H, Yap CW, Ung CY, Xue Y, Cao ZW, Chen YZ. Effect of selection of molecular descriptors on the prediction of blood−brain barrier penetrating and nonpenetrating agents by statistical learning methods. J Chem Inf Model. 2005;45(5):1376–84.CrossRefPubMedGoogle Scholar
  25. 25.
    Zhang L, Zhu H, Oprea TI, Golbraikh A, Tropsha A. QSAR modeling of the blood–brain barrier permeability for diverse organic compounds. Pharm Res. 2008;25(8):1902–14.CrossRefPubMedGoogle Scholar
  26. 26.
    Jung E, Kim J, Kim M, Jung DH, Rhee H, Shin J-M, et al. Artificial neural network models for prediction of intestinal permeability of oligopeptides. BMC Bioinformatics. 2007;8(1):245.CrossRefPubMedPubMedCentralGoogle Scholar
  27. 27.
    Garg P, Verma J. In silico prediction of blood brain barrier permeability: an artificial neural network model. J Chem Inf Model. 2006;46(1):289–97.CrossRefPubMedGoogle Scholar
  28. 28.
    Hou T, Wang J, Li Y. ADME evaluation in drug discovery. 8. The prediction of human intestinal absorption by a support vector machine. J Chem Inf Model. 2007;47(6):2408–15.CrossRefPubMedGoogle Scholar
  29. 29.
    Fröhlich H, Wegner J, Sieker F, Zell A. Kernel functions for attributed molecular graphs—a new similarity-based approach to ADME prediction in classification and regression. Mol Informatics. 2006;25(4):317–26.Google Scholar
  30. 30.
    Shen J, Cheng F, Xu Y, Li W, Tang Y. Estimation of ADME properties with substructure pattern recognition. J Chem Inf Model. 2010;50(6):1034–41.CrossRefPubMedGoogle Scholar
  31. 31.
    Golmohammadi H, Dashtbozorgi Z, Acree WE. Quantitative structure–activity relationship prediction of blood-to-brain partitioning behavior using support vector machine. Eur J Pharm Sci. 2012;47(2):421–9.CrossRefPubMedGoogle Scholar
  32. 32.
    Vapnik V. The nature of statistical learning theory. Berlin: Springer science & business media; 2013.Google Scholar
  33. 33.
    Trotter MWB. Support vector machines for drug discovery. London: University of London; 2007.Google Scholar
  34. 34.
    Geppert H, Vogt M, Bajorath J. Current trends in ligand-based virtual screening: molecular representations, data mining methods, new application areas, and performance evaluation. J Chem Inf Model. 2010;50(2):205–16.CrossRefPubMedGoogle Scholar
  35. 35.
    Collobert R, Bengio S. SVMTorch: Support vector machines for large-scale regression problems. J Mach Learn Res. 2001;1(Feb):143–60.Google Scholar
  36. 36.
    Smola AJ, Schölkopf B. A tutorial on support vector regression. Stat Comput. 2004;14(3):199–222.CrossRefGoogle Scholar
  37. 37.
    Gunn SR. Support vector machines for classification and regression. ISIS Tech Rep. 1998;14:85–6.Google Scholar
  38. 38.
    Chung K-M, Kao W-C, Sun C-L, Wang L-L, Lin C-J. Radius margin bounds for support vector machines with the RBF kernel. Neural Comput. 2003;15(11):2643–81.CrossRefPubMedGoogle Scholar
  39. 39.
    Hsu C-W, Chang C-C, Lin C-J. A practical guide to support vector classification. 2003.Google Scholar
  40. 40.
    Yap CW. PaDEL-descriptor: an open source software to calculate molecular descriptors and fingerprints. J Comput Chem. 2011;32(7):1466–74.CrossRefPubMedGoogle Scholar
  41. 41.
    Toolkits O. OpenEye Scientific Software, Santa Fe, NM. 2015.Google Scholar
  42. 42.
    Kennard RW, Stone LA. Computer aided design of experiments. Technometrics. 1969;11(1):137–48.CrossRefGoogle Scholar
  43. 43.
    Chang C-C, Lin C-J. LIBSVM: a library for support vector machines. ACM Trans Intell Syst Technol. 2011;2(3):27.CrossRefGoogle Scholar
  44. 44.
    Zheng F, Bayram E, Sumithran SP, Ayers JT, Zhan C-G, Schmitt JD, et al. QSAR modeling of mono- and bis-quaternary ammonium salts that act as antagonists at neuronal nicotinic acetylcholine receptors mediating dopamine release. Bioorg Med Chem. 2006;14:3017–37.CrossRefPubMedGoogle Scholar
  45. 45.
    Zheng F, McConnell MJ, Zhan C-G, Dwoskin LP, Crooks PA. QSAR study on maximal inhibition (Imax) of quaternary ammonium antagonists for S-(−)-nicotine-evoked dopamine release from dopaminergic nerve terminals in rat striatum. Bioorg Med Chem. 2009;17:4477–85.CrossRefPubMedPubMedCentralGoogle Scholar
  46. 46.
    Zheng F, Zheng G, Deaciuc AG, Zhan C-G, Dwoskin LP, Crooks PA. Computational neural network analysis of the affinity of lobeline and tetrabenazine analogs for the vesicular monoamine transporter-2. Bioorg Med Chem. 2007;15:2975–92.CrossRefPubMedPubMedCentralGoogle Scholar
  47. 47.
    Zheng F, Zheng G, Deaciuc AG, Zhan C-G, Dwoskin LP, Crooks PA. Computational neural network analysis of the affinity of N-n-alkylnicotinium salts for the alpha4beta2* nicotinic acetylcholine receptor. J Enzyme Inhib Med Chem. 2007;24:157–68.CrossRefGoogle Scholar
  48. 48.
    Ring JR, Zheng F, Haubner AJ, Littleton JM, Crooks PA. Improving the inhibitory activity of arylidenaminoguanidine compounds at the N-methyl-D-aspartate receptor complex from a recursive computational-experimental structure-activity relationship study. Bioorg Med Chem. 2013;21:1764–74.CrossRefPubMedGoogle Scholar
  49. 49.
    Baldi P, Brunak S, Chauvin Y, Andersen CA, Nielsen H. Assessing the accuracy of prediction algorithms for classification: an overview. Bioinformatics. 2000;16(5):412–24.CrossRefPubMedGoogle Scholar

Copyright information

© American Association of Pharmaceutical Scientists 2018

Authors and Affiliations

  1. 1.Center for Pharmaceutical Innovation and ResearchUniversity of KentuckyLexingtonUSA
  2. 2.Molecular Modeling and Biopharmaceutical CenterUniversity of KentuckyLexingtonUSA
  3. 3.Department of Pharmaceutical Sciences, College of PharmacyUniversity of KentuckyLexingtonUSA

Personalised recommendations