Publicação

Deep learning for identification of pathogenic genetic mutations

Ver documento

Detalhes bibliográficos
Resumo:Nowadays, in which the world is highly developed technologically, is possible to perform several tasks in certain areas that aim to help the world’s population. One of the areas where it has invested more in technological resources is in bioinformatics area. The growth in this area is a signal of improvement in quality of life of population, and this improvement may pass through an analysis of genetic code, alerting her of possible genetic changes or even the appearing of the diseases. This dissertation aims to perform an analysis to genetic code in order to know if an change may be pathogenic or not. In a first step, are performed tests with classical classifiers, to know their behaviour. Then, are performed new tests but this time using different models based on convolutional neural networks to get a better prediction and results of the same. Lastly, is done a comparison between each adopted classifier in order to be applied in the future the respective models in bioinformatics area.
Autores principais:Vieira, Pedro Gabriel Fernandes
Assunto:Engenharia de computadores e telemática Doenças genéticas Bioinformática Aprendizagem automática
Ano:2017
País:Portugal
Tipo de documento:dissertação de mestrado
Tipo de acesso:acesso aberto
Instituição associada:Universidade de Aveiro
Idioma:inglês
Origem:RIA - Repositório Institucional da Universidade de Aveiro
Descrição
Resumo:Nowadays, in which the world is highly developed technologically, is possible to perform several tasks in certain areas that aim to help the world’s population. One of the areas where it has invested more in technological resources is in bioinformatics area. The growth in this area is a signal of improvement in quality of life of population, and this improvement may pass through an analysis of genetic code, alerting her of possible genetic changes or even the appearing of the diseases. This dissertation aims to perform an analysis to genetic code in order to know if an change may be pathogenic or not. In a first step, are performed tests with classical classifiers, to know their behaviour. Then, are performed new tests but this time using different models based on convolutional neural networks to get a better prediction and results of the same. Lastly, is done a comparison between each adopted classifier in order to be applied in the future the respective models in bioinformatics area.