QSAR study of heparanase inhibitors activity using artificial neural networks and Levenberg-Marquardt algorithm.

A linear and non-linear quantitative structure-activity relationship (QSAR) study is presented for modeling and predicting heparanase inhibitors' activity. A data set that consisted of 92 derivatives of 2,3-dihydro-1,3-dioxo-1H-isoindole-5-carboxylic acid, furanyl-1,3-thiazol-2-yl and benzoxazol-5-yl acetic acids is used in this study. Among a large number of descriptors, four parameters classified as physico-chemical, topological and electronic indices are chosen using stepwise multiple regression technique. The artificial neural networks (ANNs) model shows superiority over the multiple linear regressions (MLR) by accounting 87.9% of the variances of antiviral potency of the heparanase inhibitors. This paper focuses on investigating the role of weight update functions in developing ANNs. Levenberg-Marquardt (L-M) algorithm shows a better performance compared with basic back propagation (BBP) and conjugate gradient (CG) algorithms. The accuracy of 4-3-1 L-M ANN model was illustrated using leave-one-out (LOO), leave-multiple-out (LMO) cross-validations and Y-randomization. The mean effect of descriptors and sensitivity analysis show that log P is the most important parameter affecting the inhibitory behavior of the molecules.

[1]  M. Götte,et al.  Functions of cell surface heparan sulfate proteoglycans. , 1999, Annual review of biochemistry.

[2]  L. Kjellén,et al.  Proteoglycans: structures and interactions. , 1991, Annual review of biochemistry.

[3]  Julio Caballero,et al.  Modeling of activity of cyclic urea HIV-1 protease inhibitors using regularized-artificial neural networks. , 2006, Bioorganic & medicinal chemistry.

[4]  Ram Sagar,et al.  CP-MLR directed QSAR studies on the antimycobacterial activity of functionalized alkenols--topological descriptors in modeling the activity. , 2005, Bioorganic & medicinal chemistry.

[5]  Julio Caballero,et al.  Linear and nonlinear QSAR study of N-hydroxy-2-[(phenylsulfonyl)amino]acetamide derivatives as matrix metalloproteinase inhibitors. , 2006, Bioorganic & medicinal chemistry.

[6]  Yizeng Liang,et al.  QSPR modeling of thermal conductivity detection response factors for diverse organic compound , 2006 .

[7]  Man-Wai Mak,et al.  A conjugate gradient learning algorithm for recurrent neural networks , 1999, Neurocomputing.

[8]  Rajni Garg,et al.  A mechanistic study of 3-aminoindazole cyclic urea HIV-1 protease inhibitors using comparative QSAR. , 2004, Bioorganic & medicinal chemistry.

[9]  Zheng Tang,et al.  A modified error function for the backpropagation algorithm , 2004, Neurocomputing.

[10]  M G Ford,et al.  Selecting compounds for focused screening using linear discriminant analysis and artificial neural networks. , 2004, Journal of molecular graphics & modelling.

[11]  A. Khalafi-nezhad,et al.  Design, synthesis, antibacterial and QSAR studies of benzimidazole and imidazole chloroaryloxyalkyl derivatives. , 2005, Bioorganic & medicinal chemistry.

[12]  W. Cowden,et al.  Identification of sulfated oligosaccharide-based inhibitors of tumor growth and metastasis using novel in vitro assays for angiogenesis and heparanase activity. , 1999, Cancer research.

[13]  M. Doble,et al.  QSAR studies of paeonol analogues for inhibition of platelet aggregation. , 2005, Bioorganic & medicinal chemistry.

[14]  I. Vlodavsky,et al.  Molecular properties and involvement of heparanase in cancer metastasis and angiogenesis. , 2001, The Journal of clinical investigation.

[15]  Kunal Roy,et al.  QSAR by LFER model of HIV protease inhibitor mannitol derivatives using FA-MLR, PCRA, and PLS techniques. , 2006, Bioorganic & medicinal chemistry.

[16]  Claude Mirodatos,et al.  Using Artificial Neural Networks to Boost High‐throughput Discovery in Heterogeneous Catalysis , 2004 .

[17]  Didier Villemin,et al.  Neural Networks: Accurate Nonlinear QSAR Model for HEPT Derivatives , 2003, J. Chem. Inf. Comput. Sci..

[18]  Mati Karelson,et al.  Six-membered cyclic ureas as HIV-1 protease inhibitors: a QSAR study based on CODESSA PRO approach. Quantitative structure-activity relationships. , 2002, Bioorganic & medicinal chemistry letters.

[19]  Maykel Pérez González,et al.  A TOPS-MODE approach to predict adenosine kinase inhibition. , 2004, Bioorganic & medicinal chemistry letters.

[20]  R. A. Ellis,et al.  Molecular modeling system. , 1972, Journal of molecular biology.

[21]  Yahya H. Zweiri,et al.  A three-term backpropagation algorithm , 2003, Neurocomputing.

[22]  Hiromi Miyajima,et al.  Predicting melting temperature (Tm) of oligoribonucleotide duplex by neural network , 2002 .

[23]  H. Díaz,et al.  A TOPS-MODE approach to predict permeability coefficients , 2004 .

[24]  W Vycudilik,et al.  Prediction of gas chromatographic retention indices of a diverse set of toxicologically relevant compounds. , 2004, Journal of chromatography. A.

[25]  Dan W. Patterson,et al.  Artificial Neural Networks: Theory and Applications , 1998 .

[26]  D. Marquardt.  An Algorithm for Least-Squares Estimation of Nonlinear Parameters , 1963 .

[27]  M. Jalali-Heravi,et al.  Artificial neural network modeling of Kováts retention indices for noncyclic and monocyclic terpenes. , 2001, Journal of chromatography. A.

[28]  J. Zupan,et al.  Neural networks: A new method for solving chemical problems or just a passing phase? , 1991 .

[29]  R. Iozzo.  Matrix proteoglycans: from molecular design to cellular function. , 1998, Annual review of biochemistry.

[30]  I. Vlodavsky,et al.  Molecular properties and involvement of heparanase in cancer progression and normal development. , 2001, Biochimie.

[31]  Roberto Todeschini,et al.  Handbook of Molecular Descriptors , 2002 .

[32]  J. Bennett,et al.  2,3-Dihydro-1,3-dioxo-1H-isoindole-5-carboxylic acid derivatives: a novel class of small molecule heparanase inhibitors. , 2004, Bioorganic & medicinal chemistry letters.

[33]  M. Nakajima,et al.  Human Heparanase , 1999, The Journal of Biological Chemistry.

[34]  Mehdi Jalali-Heravi,et al.  Use of Artificial Neural Networks in a QSAR Study of Anti-HIV Activity for a Large Group of HEPT Derivatives , 2000, J. Chem. Inf. Comput. Sci..