Thesis Details
Multi-Agent System for the Prediction of the Effect of Mutations on Protein Stability
Proteins are building blocks of every living organism, as they are responsible for multiple crucial functions. They consist of amino acids chains and these chains can be changed. The change is called mutation. Mutation can happen naturally, or created in laboratory. The~aim of this thesis is to present novel methology for determining protein's stability upon mutations. It consists of two models. The first model is multi-agent system which handles classification into two classes, i.e, stabilizing and destabilizing. The best model gained 0.7~ACC and 0.41 MCC. The second part dealt with predicting exact values of G where an Extreme Gradient Boosting model was created which managed to gain 1.67 RMSE with 0.53 PCC. New datasets for training and validation, which are truly independent, were also introduced in this thesis.
protein, machine learning, protein stability, classification, regression, mutations, Extreme Gradient Boosting
Bartík Vladimír, Ing., Ph.D. (DIFS FIT BUT), člen
Hruška Tomáš, prof. Ing., CSc. (DIFS FIT BUT), člen
Hynek Jiří, Ing., Ph.D. (DIFS FIT BUT), člen
Veselý Vladimír, Ing., Ph.D. (DIFS FIT BUT), člen
Vojnar Tomáš, prof. Ing., Ph.D. (DITS FIT BUT), člen
@mastersthesis{FITMT24374,
author = "Ond\v{r}ej Dosed\v{e}l",
type = "Master's thesis",
title = "Multi-Agent System for the Prediction of the Effect of Mutations on Protein Stability",
school = "Brno University of Technology, Faculty of Information Technology",
year = 2022,
location = "Brno, CZ",
language = "english",
url = "https://www.fit.vut.cz/study/thesis/24374/"
}