Predicting the evolutionary pathways of the SARS-COV-2 spike protein by mutability landscapes

Viruses are known as infectious agents and can cause diseases and deaths in humans. Although vaccines and drugs have been developed to prevent these effects, mutations occurring in the genetic structures of viruses may enable them to acquire characteristics such as evading the immune system and binding more effectively to host cells, thereby becoming resistant to existing treatments. The global COVID-19 pandemic has highlighted the necessity of monitoring and predicting genetic changes in the virus quickly, reliably, and effectively as new SARS-CoV-2 variants emerge. Developing new drugs and vaccines that match the virus's updated genetic content is costly and time-consuming. Therefore, predicting possible mutations in the SARS-CoV-2 genome and their potential effects before they occur, and taking the necessary precautions, is of great importance. In this thesis, the aim is to study the evolutionary process of the SARS-CoV-2 virus, which causes COVID-19 pandemic, and to predict significant future mutations. To achieve this, genetic sequences of the spike protein region, which is responsible for the virus infecting human cells, were first obtained from the GISAID database. These sequences were then aligned using various multiple sequence alignment methods (such as Clustal Omega, MAFFT, and T-Coffee). Subsequently, these alignments were used to calculate mutability scores for each amino acid based on different scoring functions (such as Karlin, Sander, Valdar, etc.). By examining the correlation between the obtained scores, regions of the virus that are either conserved or prone to change were identified. Using the calculated mutability scores and experimental mutation rates, potential future mutations were predicted through a random walk method. The studies conducted in this thesis are expected to contribute to the prediction of the evolutionary trajectory of the SARS-CoV-2.