Modelagem computacional aplicada a análise de similaridade e filogenia das proteínas

Abstract

With the increasing number of protein sequencing data, the first step in identifying this macromolecule is to compare proteins to identify their similarities with known proteins. This similarity analysis can identify evolutionary relationships between the compared spe cies. Similar sequences imply species that share a recent common ancestor. However, the sequences are not always conserved in evolution, so in this case, a comparison with the more conserved structure can be an alternative to searching for evolutionary relationships between proteins with the same function. This work analyzes the similarities of protein sequences from different species to classify them according to their evolutionary charac teristics using one-dimensional cellular automata to represent each protein as an image. We calculated the distances between the automata images using the Hamming distance. This distance measures the similarities between the cellular automata images, and we use it to analyze species evolutionary relationships. We also propose modeling using the hydropathy profile difference to analyze the differences that occurred in the structures of the SARS-CoV-2 variants throughout the COVID-19 pandemic. Our method efficiently approached species of the same animal class and variants of SARS-CoV-2 that share the N501Y mutation. Furthermore, using the difference in the hydropathy profile, we can see that the Omicron variant underwent a significant change in the RBD region that may be related to the cases of reinfection for this variant.