TRATAMIENTOS MULTIOBJETIVOS APLICADOS AL PROBLEMA DE PLEGAMIENTO DE PROTEÍNAS BAJO EL MODELO HP.

FLORES CHACÓN, MANUEL SERVANDO (2017) TRATAMIENTOS MULTIOBJETIVOS APLICADOS AL PROBLEMA DE PLEGAMIENTO DE PROTEÍNAS BAJO EL MODELO HP. Maestría thesis, UNIVERSIDAD AUTONOMA DE CHIHUAHUA.

[img]
Vista Previa
Text
Tesis-Manuel-Servando-Flores-Chacon-1.pdf

Download (2180Kb) | Vista Previa

Resumen

Proteins are linear sequences of amino acids, these must be able to fold correctly without more information than the coding of amino acids that comprise it and the interactions that are established between them, under certain physiological conditions, the proteins fold up to find their native three-dimensional conformation, or in other words the functionally active conformation. The hydrophobic-polar model is an abstraction of the behavior of proteins, based on the hydrophobic iterations of certain amino acids. The model searches by means of a mono objective energy function to determine the functional protein configuration. In the literature there are different versions of the energy function for the model, which approach the problem from different points of view. Within these functions are multiobjective alternatives, which seek to abound on the objective space in the hope that said search provide better results than the original energy function. In this work we present aspects to work with a multi-objective version of the energy function used by the model, using a multi-objective evolutionary algorithm, implementing the Hill climbing philosophy on variation operators. The experiments showed that the proposed algorithm can be used with excitation for instances of length less than or equal to 36 for a square grid in 2D, whereas for one in 3D the algorithm allowed to have favorable results for instances less than or equal to 18. By subjecting the algorithm to larger instances, the algorithm shows competitive results with respect to those shown in the literature.

Tipo de Documento: Tesis (Maestría)
Palabras Clave: Keywords: Optimization, Hill-climbing, Path Relinking, NSGA-ll, Pull-Moves, Genetic Algorithms
Clasificación temática: Q Science > QA Mathematics > QA75 Electronic computers. Computer science
Usuario Remitente: Admin Administrador del Respositorio
Depositado: 31 Oct 2017 20:09
Ultima Modificación: 07 Nov 2017 14:06
URI: http://repositorio.uach.mx/id/eprint/124

Actions (login required)

Ver Objeto Ver Objeto

Universidad Autonoma de Chihuahua