Predição e validação estrutural de macromoléculas complexas com estudo de caso envolvendo proteínas do Sars-Cov-2

Abstract

The three-dimensional structure of a protein is important because the function of the protein is linked to both its atomic composition and its three-dimensional structure, and in the case of a virus, prediction in a faster and simpler way speeds up the creation of vaccines and medicines to fight it . This dissertation presents mathematical-computational and physical-chemical aspects involved in the reconstruction of the three-dimensional molecular structure of proteins, using proteins from the SARS-CoV-2 virus as a case study. For this, the main algorithms that solve the Molecular Distance Geometry Problem (MDGP) were implemented, variations of one of the algorithms were proposed and tested, and a technical visit to the National Center for Nuclear Magnetic Resonance at UFRJ was carried out, where it was possible to analyze the method of obtaining data through Nuclear Magnetic Resonance. After analyzing the implemented methods, the need for chemical validation for the generated structures was identified, since the structural calculation only guarantees the mathematical validity of the results, so a structural reconstruction methodology was created, ranging from data search, creation from test instances to calculation and structural validation. This methodology was used in the case study carried out with proteins of the new coronavirus, mainly in the variants that reached the state of Amazonas.