Mapeamento de Parâmetros do Simulated Annealing Generalizado aplicado ao problema do Enovelamento de Proteínas

• Autor(a) principal: Agostini, Flavia Paiva
• Data de Publicação: 2008
• Tipo de documento: Tese
• Idioma: por
• Título da fonte: Biblioteca Digital de Teses e Dissertações do LNCC
• Texto Completo: https://tede.lncc.br/handle/tede/105
• topic: Proteínas-Estrutura; Otimização Matemática; Generalized Simulated Annealing (GSA); Protein - Structures; Optimization Mathematica; Generalized Simulated Annealing (GSA); CNPQ::CIENCIAS EXATAS E DA TERRA::CIENCIA DA COMPUTACAO

As the genome sequencing advances, the comprehension of protein structures becomes a crucial extension to these progresses. In spite of the numerous recent technological advances, experimental determination of protein terciary structures is still very slow compared to the accumulated data from amino acid sequences. That is what makes the protein folding a central problem to the development of the pots-genomic era. In this work we use an optimization method, the Generalized Simulated Annealing (GSA), which is based on Tsallis' generalized thermostatistics, to investigate the protein folding problem. Although GSA is a generic procedure, its efficiency depends not only on the appropriate choice of parameters, but also on topological characteristics of the energy hypersurface. By mapping all the GSA parameters, it can be possible to reduce the number of possible choices of them. That also allows an analysis of its effects on the algorithm behavior. As a initial step, we apply GSA to known structures, such as polyalanines. In sequence, we also apply GSA to three more peptides of ribosomal P proteins, which are of considerable importance on the comprehension of Chagas' heart disease. Each one contains 13 amino acids and differ only on the third residue by a non-conservative mutation. As these peptides do not have experimentally resolved structure, we analyze results obtained from GSA followed by Molecular Dynamics simulations. Validity of these results is studied such that, in the future, unknown structures can be determined by this technique with a higher degree of confidence.