Um modelo de workflow científico para o refinamento da estrutura 3D aproximada de proteínas

• Ano de defesa: 2016
• Autor(a) principal: Soletti, Leonardo Veronese
• Orientador(a): Souza, Osmar Norberto de
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Pontifícia Universidade Católica do Rio Grande do Sul
• Programa de Pós-Graduação: Programa de Pós-Graduação em Ciência da Computação
• Departamento: Faculdade de Informática
• País: Brasil
• Palavras-chave em Português: Bioinformática; Bioinformática Estrutural; Workflows Científicos; Predição da Estrutura 3D de Proteínas; Dinâmica Molecular
• Área do conhecimento CNPq: CIENCIAS EXATAS E DA TERRA::CIENCIA DA COMPUTACAO
• Link de acesso: http://tede2.pucrs.br/tede2/handle/tede/7383

As a consequence of the post-genomic era an explosion of information and numerous discoveries made available large amounts of biological data. Even with the technology enhancements regarding protein structure prediction techniques, it is still not possible to find a tool to predict with precision the exact the three-dimensional structure of a given protein. This brings new challenges, starting from how to understand and organize these resources until sharing and reuse of successful experiments, as well as how to provide interoperability between data from different sources, without mentioning the diversity between tools and different user profiles. This kind of data flow is regularly addressed as command line scripts which require users to have programming skills. Such scripts have problems interfering, collecting and storing data while executing. Furthermore, these scripts and can be very complex leading to difficulties of implementation, maintenance and reuse. Another problem that arises when a set of tasks are proposed to be conducted through scripts is the possibility of missing any step in the process or running at incorrect order, leading to inconsistent results. It becomes necessary techniques and tools to ease this process in an organized way as a sequence of steps characterized by a workflow, thus automating this process. In this context, we sought to develop a scientific workflow model using bioinformatics tools and biology expertise to automate the process of protein refinement of polypeptides predicted by CReF method once the refinement process scripts were automated, it was possible to increase the amount of experiments while maintaining an acceptable quality criteria. Finally, was developed a web interface that facilitates the visualization of the results in an organized way.