Automatic adjustment of the relative importance of different input variables for optimization of counter-propagation artificial neural networks.

In this work we present a quantitative structure-activity relationship study with 49 peptidic molecules, inhibitors of the HIV-1 protease. The modelling was preformed using counter-propagation artificial neural networks (CPANN), an algorithm which has been proven as a valuable tool for data analysis. The initial pre-processing of the data involved auto-scaling, which gives equal importance to all the variables considered in the model. In order to enhance the influence of some of the variables that carry valuable information for improvement of the model, we introduce a novel approach for adjustment of the relative importance of different input variables. Having involved a genetic algorithm, the relative importance was adjusted during the training of the CPANN. The proposed approach is capable of finding simpler efficient models, when compared to the approach with the original, i.e. equally important input variables. A simpler model also means more robust and less subjected to the overfitting model, therefore we consider the proposed procedure as a valuable improvement of the CPANN algorithm.

[1]  M Pastor,et al.  Comparative binding energy analysis of HIV-1 protease inhibitors: incorporation of solvent effects and validation as a powerful tool in receptor-based drug design. , 1998, Journal of medicinal chemistry.

[2]  M Pastor,et al.  Simulation of alternative binding modes in a structure-based QSAR study of HIV-1 protease inhibitors. , 1997, Journal of molecular graphics & modelling.

[3]  Gerhard Hessler,et al.  Drug Design Strategies for Targeting G‐Protein‐Coupled Receptors , 2002, Chembiochem : a European journal of chemical biology.

[4]  Rudolf Kiralj,et al.  A priori molecular descriptors in QSAR: a case of HIV-1 protease inhibitors. I. The chemometric approach. , 2003, Journal of molecular graphics & modelling.

[5]  Stephen Grossberg,et al.  A massively parallel architecture for a self-organizing neural pattern recognition machine , 1988, Comput. Vis. Graph. Image Process..

[6]  L. Pogliani.  Structure property relationships of amino acids and some dipeptides , 1994, Amino Acids.

[7]  Marjana Novič,et al.  Counter-propagation neural networks in Matlab , 2008 .

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

[9]  Emilio Benfenati,et al.  Modeling Toxicity by Using Supervised Kohonen Neural Networks , 2003, J. Chem. Inf. Comput. Sci..

[10]  Subhash C. Basak,et al.  Proteomic Maps-Toxicity Relationship of Halocarbons Studied with Similarity Index and Genetic Algorithm , 2006, J. Chem. Inf. Model..

[11]  Yitzhak Mendelson,et al.  Multivariate Determination of Glucose in Whole Blood Using Partial Least-Squares and Artificial Neural Networks Based on Mid-Infrared Spectroscopy , 1993 .

[12]  Remo Guidieri.  Res , 1995, RES: Anthropology and Aesthetics.

[13]  C Helma,et al.  Validation of counter propagation neural network models for predictive toxicology according to the OECD principles: a case study , 2006, SAR and QSAR in environmental research.

[14]  P Barbieri,et al.  Application of counterpropagation artificial neural network for modelling properties of fish antibiotics , 2004, SAR and QSAR in environmental research.

[15]  Gerhard Klebe,et al.  Recent developments in structure-based drug design , 2000, Journal of Molecular Medicine.

[16]  Károly Héberger,et al.  Prediction of Anti-HIV-1 Activity of a Series of Tetrapyrrole Molecules , 2003, J. Chem. Inf. Comput. Sci..

[17]  J. Hermens,et al.  Classifying environmental pollutants , 1992 .

[18]  H. van de Waterbeemd,et al.  ADMET in silico modelling: towards prediction paradise? , 2003, Nature reviews. Drug discovery.

[19]  Jure Zupan,et al.  Study of structure–toxicity relationship by a counterpropagation neural network , 1999 .

[20]  Marjana Novic,et al.  Variable Selection and Interpretation in Structure-Affinity Correlation Modeling of Estrogen Receptor Binders , 2005, J. Chem. Inf. Model..

[21]  Stephen Grossberg,et al.  Nonlinear neural networks: Principles, mechanisms, and architectures , 1988, Neural Networks.

[22]  T. Halgren,et al.  A priori prediction of activity for HIV-1 protease inhibitors employing energy minimization in the active site. , 1995, Journal of medicinal chemistry.

[23]  Igor Kuzmanovski,et al.  Optimization of artificial neural networks for prediction of the unit cell parameters in orthorhombic perovskites. Comparison with multiple linear regression , 2003 .

[24]  R. Leardi,et al.  Genetic algorithms applied to feature selection in PLS regression: how and when to use them , 1998 .