Mejías CY, Carrasco VR, Ochoa II, Moreno LE

Language: Spanish
References: 14
Page: 1-19
PDF size: 266.68 Kb.

ABSTRACT

The multilayer perceptron (PMC) ranks among the types of artificial neural networks (ANN), which has provided better results in studies of structure-activity relationship. As the data volumes in Bioinformatics’ projects are eventually big, it was proposed to evaluate algorithms to shorten the training time of the network without affecting its efficiency. There were evaluated different tools that work with ANN and were selected Weka algorithm for extracting the network and the Platform for Distributed Task Tarenal to distribute the training of multilayer perceptron. Finally, it was developed a training algorithm for local and distributed the MLP with the possibility of varying transfer functions. It was shown that depending on the training sample, the change of transfer functions can yield results much more efficient than those obtained with the classic sigmoid function with increased g-media between 4.5 and 17 %. Moreover, it was found that with distributed training can be achieved eventually, better results than those achieved in the local environment.

REFERENCES

1. Sremac S, Skrbk B, Onjia A. Artificial neural network prediction of quantitative structure - retention relationships of polycyclic aromatic hydocarbons in gas chromatography. J Serb Chem Soc. 2005;70(11):1291-300.

2. Bolanca T, Cerjan-Stefanovic S, Ukic S, Rogosic M, Lusa M. Application of different training methodologies for the development of a back propagation artificial neural network retention model in ion chromatography. J Chemometrics. 2008;22:106-13.

3. Lendasse A, Ji Y, Reyhani N, Verleysen M. LS-SVM Hyperparameter Selection with a Nonparametric Noise Estimator. In: Artificial Neural Networks: Formal Models and Their Applications–ICANN 2005. Springer Berlin Heidelberg; 2005:625-30.

4. Sutherland JJ, O’Brien LA, Weaver DF. A Comparison of Methods for Modeling Quantitative Structure-Activity Relationships. J Med Chem. 2004;47(22):5541-54.

5. Lucic B, Nadramija D, Basic I, Trinajstic N. Toward Generating Simpler QSAR Models: Nonlinear Multivariate Regression versus Several Neural Network Ensembles and Some Related Methods. J Chem Inf Comput Sci. 2003;43:1094-102.

6. Huang L, Lu H-M, Dai Y. Feature Selection of Support Vector Regression for Quantitative Structure-Activity Relationships (QSAR). The International Conference on Mathematics and Engineering Techniques in Medicine and Biological Sciences (METMBS’03) (Paper ID # : 269ME); 2003.

7. Escalona-Arranz JC, Carrasco-Velar R, Padrón-García JA. Introducción al diseño racional de fármacos. La Habana: Editorial Universitaria, Ministerio de Educación Superior; 2008.

8. Hall M, Frank E, Holmes G, Pfahringer B, Reutemann P, Witten IH. The WEKA Data Mining Software: An Update. ACM SIGKDD Explorations Newsletter. 2009;11(1):10-18.

9. Colectivo de Autores. Tarenal, Plataforma de Tareas Distribuidas, Proyecto Grid. Manual del desarrollador. Universidad de las Ciencias Informáticas, La Habana, Cuba; 2009.

10. SPSS Inc. SPSS 13 for Windows. 13.0 ed: Copyrigth (c) SPSS Inc. 1989-2004. All rigths reserved; 2004.

11. Universidad Complutense de Madrid. Departamento de Sociología. Introducción al análisis de datos [citado el 15 Marzo 2013] Disponible en: http://www.ucm.es/info/socivmyt/paginas/D_departamento/materiales/ analisis_datosyMultivariable/19nparam_SPSS.pdf

12. Heaton J. Introduction to Neural Networks for Java. 2nd ed. St. Louis, Missouri: Heaton Research, Inc.; 2008.

13. Carrasco-Velar R. Nuevos descriptores atómicos y moleculares para estudios de estructura-actividad: Aplicaciones [PhD. Thesis]. La Habana: Universidad de La Habana; 2008.

14. Hernández-Díaz Y. Desarrollo de modelos de clasificación de actividad biológica empleando Máquinas de Soporte Vectorial [Master]. La Habana: Universidad de las Ciencias Informáticas; 2010.