Abstract
Purpose
Predicting human skin permeability of chemical compounds accurately and efficiently is useful for developing dermatological medicines and cosmetics. However, previous work have two problems; 1) quality of databases used, and 2) methods for prediction models. In this paper, we attempt to solve these two problems.
Methods
We first compile, by carefully screening from the literature, a novel dataset of chemical compounds with permeability coefficients, measured under consistent experimental conditions. We then apply machine learning techniques such as support vector regression (SVR) and random forest (RF) to our database to develop prediction models. Molecular descriptors are fully computationally obtained, and greedy stepwise selection is employed for descriptor selection. Prediction models are internally and externally validated.
Results
We generated an original, new database on human skin permeability of 211 different compounds from aqueous donors. Nonlinear SVR achieved the best performance among linear SVR, nonlinear SVR, and RF. The determination coefficient, root mean square error, and mean absolute error of nonlinear SVR in external validation were 0.910, 0.342, and 0.282, respectively.
Conclusions
We provided one of the largest datasets with purely experimental log k p and developed reliable and accurate prediction models for screening active ingredients and seeking unsynthesized compounds of dermatological medicines and cosmetics.
Similar content being viewed by others
Abbreviations
- ALOGP:
-
Ghose–Crippen octanol–water partition coefficient
- ANN:
-
Artificial neural network
- C d :
-
Chemical concentration in dose formulation
- J ss :
-
Steady state flux of the solute
- K :
-
Skin–vehicle partition coefficient
- k p :
-
Permeability coefficient
- L :
-
Thickness of the skin
- log P:
-
Octanol–water partition coefficient
- MAE:
-
Mean absolute error
- MW:
-
Molecular weight
- PCA:
-
Principal component analysis
- QSPR:
-
Quantitative structure–property relationship
- R 2 :
-
Determination coefficient
- RF:
-
Random forest
- RMSE:
-
Root mean square error
- SVR:
-
Support vector regression
- SVR-G:
-
Support vector regression with Gaussian (radial basis function) kernel
- SVR-L:
-
Support vector regression with linear kernel
References
Bartek MJ, LaBudde JA, Maibach HI. Skin permeability in vivo: comparison in rat, rabbit, pig and man. J Investig Dermatol. 1972;58(3):114–23.
Franz TJ. Percutaneous absorption on the relevance of in vitro data. J Investig Dermatol. 1975;64(3):190–5.
Zhang Q, Grice JE, Li P, Jepps OG, Wang GJ, Roberts MS. Skin solubility determines maximum transepidermal flux for similar size molecules. Pharm Res. 2009;26(8):1974–85.
Takeuchi H, Ishida M, Furuya A, Todo H, Urano H, Sugibayashi K. Influence of skin thickness on the in vitro permeabilities of drugs through Sprague-Dawley rat or Yucatan micropig skin. Biol Pharm Bull. 2012;35(2):192–202.
Karadzovska D, Riviere JE. Assessing vehicle effects on skin absorption using artificial membrane assays. Eur J Pharm Sci. 2013;50(5):569–76.
Blank IH, McAuliffe DJ. Penetration of benzene through human skin. J Investig Dermatol. 1985;85(6):522–6.
Flynn GL. Physicochemical determinants of skin absorption. In: Gerrity TR, Henry CJ, editors. Principles of route-to-route extrapolation for risk assessment. 1st ed. New York: Elsevier; 1990. p. 93–127.
Wilschut A, ten Berge WF, Robinson PJ, McKone TE. Estimating skin permeation. The validation of five mathematical skin permeation models. Chemosphere. 1995;30(7):1275–96.
Kirchner LA, Moody RP, Doyle E, Bose R, Jeffery J, Chu I. The prediction of skin permeability by using physicochemical data. ATLA. 1997;25:359–70.
Patel H, ten Berge W, Cronin MT. Quantitative structure-activity relationships (QSARs) for the prediction of skin permeation of exogenous chemicals. Chemosphere. 2002;48(6):603–13.
Chauhan P, Shakya M. Role of physicochemical properties in the estimation of skin permeability: in vitro data assessment by partial least-squares regression. SAR QSAR Environ Res. 2010;21(5–6):481–94.
Khajeh A, Modarress H. Linear and nonlinear quantitative structure-property relationship modelling of skin permeability. SAR QSAR Environ Res. 2014;25(1):35–50.
Moss GP, Sun Y, Wilkinson SC, Davey N, Adams R, Martin GP, et al. The application and limitations of mathematical modelling in the prediction of permeability across mammalian skin and polydimethylsiloxane membranes. J Pharm Pharmacol. 2011;63(11):1411–27.
Vecchia BE, Bunge AL. Skin absorption databases and predictive equations. In: Guy R, Hadgraft J, editors. Transdermal drug delivery. 2nd ed. New York: Marcel Dekker; 2003. p. 57–141.
Roberts MS, Pugh WJ, Hadgraft J, Watkinson AC. Epidermal permeability-penetrant structure relationships: 1. An analysis of methods of predicting penetration of monofunctional solutes from aqueous solutions. Int J Pharm. 1995;126(1–2):219–33.
Ghafourian T, Fooladi S. The effect of structural QSAR parameters on skin penetration. Int J Pharm. 2001;217(1–2):1–11.
Panchagnula R, Stemmer K, Ritschel WA. Animal models for transdermal drug delivery. Methods Find Exp Clin Pharmacol. 1997;19(5):335–41.
Lehman PA, Raney SG, Franz TJ. Percutaneous absorption in man in vitro-in vivo correlation. Skin Pharmacol Physiol. 2011;24(4):224–30.
Potts RO, Guy RH. Predicting skin permeability. Pharm Res. 1992;9(5):663–9.
Lim CW, Fujiwara S, Yamashita F, Hashida M. Prediction of human skin permeability using a combination of molecular orbital calculations and artificial neural network. Biol Pharm Bull. 2002;25(3):361–6.
Katritzky AR, Dobchev DA, Fara DC, Hür E, Tämm K, Kurunczi L, et al. Skin permeation rate as a function of chemical structure. J Med Chem. 2006;49(11):3305–14.
Chen LJ, Lian GP, Han LJ. Prediction of human skin permeability using artificial neural network (ANN) modeling. Acta Pharmacol Sin. 2007;28(4):591–600.
Patel J. Science of the science, drug discovery and artificial neural networks. Curr Drug Discov Technol. 2013;10(1):2–7.
Castillo E, Fontenla-Romero O, Guijarro-Berdiñas B, Alonso-Betanzos A. A global optimum approach for one-layer neural networks. Neural Comput. 2002;14(6):1429–49.
El-Sebakhy EA. Forecasting PVT properties of crude oil systems based on support vector machines modeling scheme. J Petrol Sci Eng. 2009;64(1–4):25–34.
Smola AJ, Schölkopf B. A tutorial on support vector regression. Stat Comput. 2004;14(3):199–222.
Breiman L. Random forests. Mach Learn. 2001;45(1):5–32.
Wang Y, Zheng M, Xiao J, Lu Y, Wang F, Lu J, et al. Using support vector regression coupled with the genetic algorithm for predicting acute toxicity to the fathead minnow. SAR QSAR Environ Res. 2010;21(5–6):559–70.
Yap CW, Li ZR, Chen YZ. Quantitative structure-pharmacokinetic relationships for drug clearance by using statistical learning methods. J Mol Graph Model. 2006;24(5):383–95.
Chu A, Ahn H, Halwan B, Kalmin B, Artifon EL, Barkun A, et al. A decision support system to facilitate management of patients with acute gastrointestinal bleeding. Artif Intell Med. 2008;42(3):247–59.
Monte-Moreno E. Non-invasive estimate of blood glucose and blood pressure from a photoplethysmograph by means of machine learning techniques. Artif Intell Med. 2011;53(2):127–38.
Hsieh CH, Lu RH, Lee NH, Chiu WT, Hsu MH, Li YC. Novel solutions for an old disease: diagnosis of acute appendicitis with random forest, support vector machines, and artificial neural networks. Surgery. 2011;149(1):87–93.
Robert GP, Yang W. Density-functional theory of atoms and molecules. Oxford: Oxford University Press; 1989.
Schmidt MW, Baldridge KK, Boatz JA, Elbert ST, Gordon MS, Jensen JH, et al. General atomic and molecular electronic structure system. J Comput Chem. 1993;14(11):1347–63.
Gordon MS, Schmidt MW. Advances in electronic structure theory: GAMESS a decade later. In: Dykstra CE, Frenking G, Kim KS, Scuseria GE, editors. Theory and applications of computational chemistry: the first forty years. Amsterdam: Elsevier; 2005. p. 1167–89.
Lee C, Yang W, Parr RG. Development of the Colle-Salvetti correlation-energy formula into a functional of the electron density. Phys Rev B Condens Matter. 1988;37(2):785–9.
Beck AD. A new mixing of Hartree-Fock and local density‐functional theories. J Chem Phys. 1993;98:1372–7.
R Core Team. R: A language and environment for statistical computing. R Foundation for Statistical Computing. Vienna, Austria. 2013. Available from http://www.R-project.org/.
Cortes C, Vapnik V. Support-vector networks. Mach Learn. 1995;20(3):273–97.
Vapnik V. Statistical learning theory. New York: Wiley; 1998.
Meyer D, Dimitriadou E, Hornik K, Weingessel A, Leisch F. e1071: Misc Functions of the Department of Statistics (e1071), TU Wien. R package version 1.6–2. 2014. Available from http://CRAN.R-project.org/package=e1071/.
Svetnik V, Liaw A, Tong C, Culberson JC, Sheridan RP, Feuston BP. Random forest: a classification and regression tool for compound classification and QSAR modeling. J Chem Inf Comput Sci. 2003;43(6):1947–58.
Liaw A, Wiener M. Classification and regression by randomForest. R News. 2002;2:18–22.
Burman P. A Comparative study of ordinary cross-validation, v-fold cross-validation and the repeated learning-testing methods. Biometrika. 1989;76(3):503–14.
Shao J. Linear model selection by cross-validation. J Am Stat Assoc. 1993;88(442):486–94.
Zhang P. Model selection via multifold cross validation. Ann Stat. 1993;21(1):486–94.
Golbraikh A, Tropsha A. Beware of q2! J Mol Graph Model. 2002;20(4):269–76.
Roy PP, Roy K. On some aspects of variable selection for partial least squares regression models. QSAR Comb Sci. 2008;27(3):302–13.
Abraham MH, Martins F, Mitchell RC. Algorithms for skin permeability using hydrogen bond descriptors: the problem of steroids. J Pharm Pharmacol. 1997;49(9):858–65.
Neumann D, Kohlbacher O, Merkwirth C, Lengauer T. A fully computational model for predicting percutaneous drug absorption. J Chem Inf Model. 2006;46(1):424–9.
Neely BJ, Madihally SV, Robinson Jr RL, Gasem KA. Nonlinear quantitative structure-property relationship modeling of skin permeation coefficient. J Pharm Sci. 2009;98(11):4069–84.
Law V, Knox C, Djoumbou Y, Jewison T, Guo AC, Liu Y, et al. DrugBank 4.0: shedding new light on drug metabolism. Nucleic Acids Res. 2014;42:D1091–7.
We have obtained structures of 6213 compounds in DrugBank: http://www.drugbank.ca/downloads#structures/.
Bos JD, Meinardi MM. The 500 Dalton rule for the skin penetration of chemical compounds and drugs. Exp Dermatol. 2000;9(3):165–9.
Yano T, Nakagawa A, Tsuji M, Noda K. Skin permeability of various non-steroidal anti-inflammatory drugs in man. Life Sci. 1986;39(12):1043–50.
Flynn GL, Yalkowsky SH. Correlation and prediction of mass transport across membranes. I. Influence of alkyl chain length on flux-determining properties of barrier and diffusant. J Pharm Sci. 1972;61(6):838–52.
González MP, Terán C, Teijeira M, Helguera AM. Quantitative structure activity relationships as useful tools for the design of new adenosine receptor ligands. 1. Agonist. Curr Med Chem. 2006;13(19):2253–66.
Leardi R, Boggia R, Terrile M. Genetic algorithms as a strategy for feature selection. J Chemom. 1992;6(5):267–81.
Shamsipur M, Zare-Shahabadi V, Hemmateenejad B, Akhond M. An efficient variable selection method based on the use of external memory in ant colony optimization. Application to QSAR/QSPR studies. Anal Chim Acta. 2009;646(1–2):39–46.
Lin WQ, Jiang JH, Shen Q, Shen GL, Yu RQ. Optimized block-wise variable combination by particle swarm optimization for partial least squares modeling in quantitative structure-activity relationship studies. J Chem Inf Model. 2005;45(2):486–93.
Barratt MD. Quantitative structure-activity relationships for skin permeability. Toxicol in Vitro. 1995;9(1):27–37.
Kasting GB, Smith RL, Cooper ER. Effect of lipid solubility and molecular size on percutaneous absorption. In: Shroot B, Schaefer H, editors. Skin pharmacokinetics. Basel: Kargar; 1987. p. 138–53.
Cronin MT, Dearden JC, Moss GP, Murray-Dickson G. Investigation of the mechanism of flux across human skin in vitro by quantitative structure-permeability relationships. Eur J Pharm Sci. 1999;7(4):325–30.
Buchwald P, Bodor N. A simple, predictive, structure-based skin permeability model. J Pharm Pharmacol. 2001;53(9):1087–98.
Basak SC, Mills D, Mumtaz MM. A quantitative structure-activity relationship (QSAR) study of dermal absorption using theoretical molecular descriptors. SAR QSAR Environ Res. 2007;18(1–2):45–55.
Author information
Authors and Affiliations
Corresponding author
Electronic supplementary material
Below is the link to the electronic supplementary material.
ESM 1
(DOCX 63 kb)
Rights and permissions
About this article
Cite this article
Baba, H., Takahara, Ji. & Mamitsuka, H. In Silico Predictions of Human Skin Permeability using Nonlinear Quantitative Structure–Property Relationship Models. Pharm Res 32, 2360–2371 (2015). https://doi.org/10.1007/s11095-015-1629-y
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11095-015-1629-y