Classification of Covid-19 and Covid Variants using DNA Sequence Using Short Time Fourier Transform

Abstract

Coronavirus(Covid-19) is a severe pandemic that has affected millions around the world. It was an unforeseen viral infection that happened in the year 2020 in the Wuhan city of China. The fast spread of this danger affected the health of people across the globe. In the previous twenty years, many communicated diseases are marked like SARS-CoV, severe acute respiratory syndrome coronavirus in 2002-2003 and recently middle east respiratory syndrome coronavirus (MERS-CoV) which first existed in 2012. With the spread of Covid-19, various mutations (when the basic structure of any species changes itself) were introduced, these mutations were able to infect a person with serious illness. Some of the variants were highly transferable making the variants more infectious. In this study we have tried to differentiate between MERS-COV and SARS-CoV and we have used the dataset of strain XBB.1.5, BA.5.2.1 with the use of genome datasets from NCBI for making detection procedures smooth and increasing the accuracy of virus detection in quick time. The selected datasets contain 20 genomic sequences for Sars and mers each whereas these variants contain 15 datasets of each respectively. Then the classification between sars-cov - mers-cov and between the strains XBB.1.5, BA.5.2.1 is being performed using a short fourier transform in MATLAB. The results obtained are in the form of threshold values from which we can differentiate between the sars- mers and between the strains of covid-19( XBB.1.5, BA.5.2.1).