Development of a new methodology to predict and engineer thermostable proteins

Studies of protein thermostability change prediction due to single mutations has been of great interest because of its importance for industries, biotechnological and biomedical research. However, the search of a highly accurate predictor has been unsuccessful. In this thesis, we have developed four...

Descripción completa

Detalles Bibliográficos
Autor: Robles Martín, Ana
Tipo de recurso: tesis de maestría
Fecha de publicación:2020
País:España
Institución:Universitat Autònoma de Barcelona
Repositorio:Dipòsit Digital de Documents de la UAB
Idioma:inglés
OAI Identifier:oai:ddd.uab.cat:237829
Acceso en línea:https://ddd.uab.cat/record/237829
Access Level:acceso abierto
Palabra clave:Protein stability
Computational tools and databases
Machine learning
Metapredictor
Thermostability predictor
Integrated prediction
Estabilitat de les proteïnes
Eines computacionals i bases de dades
Aprenentatge automàtic, Metapredictor
Predictor de termostabilitat
Predicció integrada
Descripción
Sumario:Studies of protein thermostability change prediction due to single mutations has been of great interest because of its importance for industries, biotechnological and biomedical research. However, the search of a highly accurate predictor has been unsuccessful. In this thesis, we have developed four metapredictors, from which RF-Classifier, and its combined model with RF-Regressor, has the best performance. Its main goal is to take advantage of all the other predictors' strengths and minimize their weaknesses Besides, we have conducted three retrospective studies and a prospective one with an alpha/beta hydrolase to filter and select the most stabilizing mutations.