Modeling the activity of furin inhibitors using artificial neural network.

Quantitative structure-activity relationship (QSAR) models were constructed for predicting the inhibition of furin-dependent processing of anthrax protective antigen of substituted guanidinylated aryl 2,5-dideoxystreptamines. Molecular descriptors calculated by E-Dragon and RECON were subjected to variable reduction using the Unsupervised Forward Selection (UFS) algorithm. The variables were then used as input for QSAR model generation using partial least squares and back-propagation neural network. Prediction was performed via a two-step approach: (i) perform classification to determine whether the molecule is active or inactive, (ii) develop a QSAR regression model of active molecules. Both classification and regression models yielded good results with RECON providing higher accuracy than that of E-DRAGON descriptors. The performance of the regression model using E-Dragon and RECON descriptors provided a correlation coefficient of 0.807 and 0.923 and root mean square error of 0.666 and 0.304, respectively. Interestingly, it was observed that appropriate representations of the protonation states of the molecules were crucial for good prediction performance, which coincides with the fact that the inhibitors interact with furin via electrostatic forces. The results provide good prospect of using the proposed QSAR models for the rational design of novel therapeutic furin inhibitors toward anthrax and furin-dependent diseases.

[1]  Martyn G. Ford,et al.  Unsupervised Forward Selection: A Method for Eliminating Redundant Variables , 2000, J. Chem. Inf. Comput. Sci..

[2]  V. Tantishaiyakul,et al.  Prediction of Pgp–ATPase interaction and rhodamine 123 efflux inhibitory activities of propafenone analogs using PLS statistics , 2005 .

[3]  Mark J. Embrechts,et al.  New developments in PEST shape/property hybrid descriptors , 2003, J. Comput. Aided Mol. Des..

[4]  E. V. Thomas,et al.  Partial least-squares methods for spectral analyses. 1. Relation to other quantitative calibration methods and the extraction of qualitative information , 1988 .

[5]  Paolo Visca,et al.  Anthrax toxin: a tripartite lethal combination1 , 2002 .

[6]  John A. Young,et al.  Anthrax toxin: receptor binding, internalization, pore formation, and translocation. , 2007, Annual review of biochemistry.

[7]  George R. Famini,et al.  THEORETICAL ALTERNATIVES TO LINEAR SOLVATION ENERGY RELATIONSHIPS , 1998 .

[8]  Igor V. Tetko,et al.  Virtual Computational Chemistry Laboratory – Design and Description , 2005, J. Comput. Aided Mol. Des..

[9]  Virapong Prachayasittikul,et al.  Quantitative prediction of imprinting factor of molecularly imprinted polymers by artificial neural network , 2005, J. Comput. Aided Mol. Des..

[10]  Rajarshi Guha,et al.  Determining the Validity of a QSAR Model - A Classification Approach , 2005, J. Chem. Inf. Model..

[11]  I. Lindberg,et al.  Protection against Anthrax Toxemia by Hexa-d-Arginine In Vitro and In Vivo , 2004, Infection and Immunity.

[12]  Kaushal Rege,et al.  High-throughput screening and quantitative structure-efficacy relationship models of potential displacer molecules for ion-exchange systems. , 2002, Biotechnology and bioengineering.

[13]  Curt M. Breneman,et al.  Electron Density Modeling of Large Systems Using the Transferable Atom Equivalent Method , 1995, Comput. Chem..

[14]  C. Breneman,et al.  QSPR analysis of HPLC column capacity factors for a set of high‐energy materials using electronic van der waals surface property descriptors computed by transferable atom equivalent method , 1997 .

[15]  M. Mock,et al.  Toxins of Bacillus anthracis. , 2001, Toxicon : official journal of the International Society on Toxinology.

[16]  Chartchalerm Isarankura-Na-Ayudhya,et al.  Prediction of GFP spectral properties using artificial neural network , 2007, J. Comput. Chem..

[17]  Maykel Pérez González,et al.  Quantitative structure-activity relationship to predict differential inhibition of aldose reductase by flavonoid compounds. , 2005, Bioorganic & medicinal chemistry.

[18]  M. Fernández,et al.  Quantitative Structure–Activity Relationship Modeling of Growth Hormone Secretagogues Agonist Activity of some Tetrahydroisoquinoline 1‐Carboxamides , 2007, Chemical biology & drug design.

[19]  Johann Gasteiger,et al.  Neural networks in chemistry and drug design , 1999 .

[20]  G. Bolt,et al.  The role of subtilisin-like proprotein convertases for cleavage of the measles virus fusion glycoprotein in different cell types. , 1998, Virology.

[21]  Chartchalerm Isarankura-Na-Ayudhya,et al.  Quantitative structure-imprinting factor relationship of molecularly imprinted polymers. , 2007, Biosensors & bioelectronics.

[22]  A. J. Duke,et al.  Quantum topology of molecular charge distributions. 1 , 1979 .

[23]  T. Komiyama,et al.  Engineered eglin c variants inhibit yeast and human proprotein processing proteases, Kex2 and furin. , 2000, Biochemistry.

[24]  M. Fernández,et al.  QSAR for non-nucleoside inhibitors of HIV-1 reverse transcriptase. , 2006, Bioorganic & medicinal chemistry.

[25]  Ian H. Witten,et al.  Data mining: practical machine learning tools and techniques, 3rd Edition , 1999 .

[26]  Barry K. Lavine,et al.  Electronic van der Waals Surface Property Descriptors and Genetic Algorithms for Developing Structure-Activity Correlations in Olfactory Databases , 2003, J. Chem. Inf. Comput. Sci..

[27]  Erik Johansson,et al.  On the selection of the training set in environmental QSAR analysis when compounds are clustered , 2000 .

[28]  C. Chi,et al.  Two engineered eglin c mutants potently and selectively inhibiting kexin or furin , 2004, FEBS letters.

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